transcriptional profiling of arabidopsis thaliana

TRANSCRIPTIONAL PROFILING OF ARABIDOPSIS THALIANA RESPONSES TO

TNT AND RDX: A ROUTE TO PLANT IMPROVEMENT FOR

PHYTOREMEDIATION

DREW ROBERT EKMAN

(Under the Direction of Jeffrey F.D. Dean)

ABSTRACT

Serial Analysis of Gene Expression (SAGE) was used to profile transcript levels

in Arabidopsis thaliana roots and assess their responses to 2,4,6-trinitrotoluene (TNT)

and 1,3,5-hexahydro-1,3,5-triazine (RDX) exposure. SAGE libraries representing control,

TNT, and RDX-exposed seedling root transcripts were constructed, and each was

sequenced to a depth of roughly 30,000 tags. More than 19,000 unique tags were

identified overall. Upon TNT exposure, the second most highly induced tag (27-fold

increase) represented a glutathione S-transferase. Several cytochrome P450 enzymes, as

well as an ABC transporter, were also highly induced by TNT exposure. These analyses

also revealed an oxidative stress response upon TNT exposure. RDX exposure also

induced a number of transcripts including: molecular chaperones and transcription

factors, as well as vacuolar proteins and peroxidases. Comparison of these transcripts to

those induced by TNT, identified significant differences, suggesting drastically different

mechanisms for the metabolism of these two compounds. Preliminary DNA microarray

analyses were also used to verify the expression levels of transcripts affected by TNT as

identified previously by SAGE. Identification of transcriptome-level responses to TNT

and RDX exposure will better define the metabolic pathways plants use to detoxify these

xenobiotic compounds, which should help us improve phytoremediation strategies

directed at these and other nitroaromatic compounds.

INDEX WORDS: Phytoremediation, Serial Analysis of Gene Expression, Microarray, Munitions,Explosives, Bioremediation, SAGE

PHYTOREMEDIATION

DREW ROBERT EKMAN

B.S., The University of Georgia, 1997

M.S., The University of Tennessee, Knoxville, 2000

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

Drew Robert Ekman

PHYTOREMEDIATION

DREW ROBERT EKMAN

Major Professor: Jeff Dean

Committee: Alan Przybyla Claiborne Glover Russell Malmberg Scott Merkle

Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia August 2003

DEDICATION

To Lis

ACKNOWLEDGEMENTS

There are several to whom I owe a sincere debt of gratitude. First and foremost I

would like to thank my major professor Jeff Dean and my EPA sponsor Lee Wolfe, both

of whom were constant sources of help and encouragement. Their mentorship was

instrumental in the successful completion of this work. I’d also like to thank Alan

Przybyla and Walt Lorenz who in addition to sharing their keen scientific insight and

technical knowledge to problems I encountered also shared their genuine enthusiasm for

science and maybe most importantly their good humor. Claiborne Glover, Scott Merkle,

Alan Przybyla, and Russell Malmberg were kind enough to serve on my committee and

for that I am grateful. Their wisdom and critical thinking served me well. Sincere thanks

also to Macarthur Long at the EPA who not only provided guidance and support during

my tenure with the EPA, but in addition to Lee, ensured that I always had what I needed

to complete my work. I owe a debt of gratitude to Steve McCutcheon also at the EPA for

being my initial link with the NNEMS (National Network for Environmental

Management Studies) fellowship program and for his continued support of the project.

That being said, I’d also like to thank the NNEMS program and the U.S. EPA for

supporting me for the majority of my time spent pursuing this degree. Deepest thanks

also to my family who supported me throughout my years in graduate school and who

always displayed genuine interest in my work. Thanks also to Katie Ekman, a canine of

14 years who was a constant fixture at my feet during the writing of this dissertation.

And finally, to my beautiful wife Lisa, who has been there for the entirety of both of our

undergraduate and graduate degrees, sharing many late nights up studying. Thank you

for being the incredible that person you are. Thank you for everything.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS.................................................................................................v

CHAPTER

1 INTRODUCTION AND LITERATURE REVIEW .........................................1

EXPLOSIVES...............................................................................................2

TOXICITY ....................................................................................................4

ENVIRONMENTAL CONTAMINATION BY EXPLOSIVES AND

METHODS OF REMOVAL...................................................................6

BIOREMEDIATION ....................................................................................8

BIOREMEDIATION OF TNT AND RDX ..................................................9

PHYTOREMEDIATION............................................................................12

PLANT METABOLISM OF TNT AND RDX...........................................15

GENOMICS ................................................................................................20

TRANSCRIPTIONAL PROFILING ..........................................................21

PHYTOREMEDIATION AND GENOMICS ............................................29

REFERENCES............................................................................................32

2 SAGE ANALYSIS OF TRANSCRIPTOME RESPONSES IN ARABIDOPSIS

THALIANA ROOTS EXPOSED TO TNT ..................................................49

ABSTRACT ................................................................................................50

INTRODUCTION.......................................................................................51

RESULTS....................................................................................................52

DISCUSSION .............................................................................................56

EXPERIMENTAL PROTOCOL ................................................................63

REFERENCES............................................................................................69

3 EFFECTS OF THE EXPLOSIVE RDX ON TRANSCRIPT EXPRESSION

IN ARABIDOPSIS THALIANA SEEDLING ROOTS.................................82

ABSTRACT ................................................................................................83

INTRODUCTION.......................................................................................83

MATERIALS AND METHODS ................................................................85

RESULTS AND DISCUSSION .................................................................90

CONCLUSIONS .........................................................................................96

REFERENCES..........................................................................................105

4 MICROARRAY ANALYSIS OF ARABIDOPSIS TRANSCRIPTOME

CHANGES INDUCED BY TNT..............................................................110

INTRODUCTION.....................................................................................111

MATERIALS AND METHODS ..............................................................113

RESULTS AND DISCUSSION ...............................................................118

REFERENCES..........................................................................................126

5 CONCLUSIONS............................................................................................128

REFERENCES..........................................................................................135

CHAPTER 1

INTRODUCTION AND LITERATURE REVIEW

2,4,6-trinitrotoluene (TNT) and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) are

compounds widely used as explosives by the military. The great utility of these

compounds is a result of their high stability (they are relatively insensititive to the

addition of heat, flame, spark, or shock) and the characteristically large amounts of

energy released upon their detonation. During the First World War, little consideration

for the potential toxicity of these compounds to the environment was exercised.

However, during World War II, when the scale of production increased dramatically,

concern over the toxicity of these compounds emerged. Unfortunately by this time,

extensive contamination of land and water surrounding the sites of their manufacture,

storage, and disposal had already occurred. Research to assess the toxicity of these

compounds to humans and other organisms soon revealed the need to remove

contaminating explosives from the environment. In response, large-scale methods of

remediation were developed. However, these methods proved environmentally

destructive, laborious, and costly. For this reason, alternative methods of remediation are

being sought. Alternatives using plants (phytoremediation) have been shown to remove

explosives effectively from contaminated environments with the advantage of producing

little impact on the surrounding landscape.

EXPLOSIVES

In contrast to primary explosives and low explosives, both of which are easily

detonated by the addition of heat, flame, spark, or shock, TNT and RDX are considered

high explosives due to their relative insensitivity to these conditions. It is through the

detonation of primary or low explosives (e.g., gunpowder) that high explosives are

detonated to release large amounts of energy. Although TNT and RDX are both often

classified as nitroaromatic explosives, TNT is the only one of these to possess true

aromaticity (Figure 1.1). Originally synthesized by Wilbrand in 1863, TNT is produced

by the sequential nitration of toluene to nitrotoluene, dinitrotoluene (DNT), and finally

TNT, using a mixture of nitric and sulfuric acids (Palmer et al., 1996). The relatively

high solubility of TNT in water (~156 mg/L at 250C) leads to the production of large

volumes of contaminated wastewater during TNT production and munitions loading

operations. In the past, this contaminated water was collected in holding lagoons, or in

some cases released directly into surface waters. Current contaminated sites typically

contain average concentrations of 10,000 mg/kg TNT in soil and 156 mg/L TNT in water

(Fernando et al., 1990).

Used as the main component in various explosive compositions for artillery shell

charges or plastic explosives, RDX (an acronym for Royal Demolition EXplosive or

Research Department EXplosive) is technically a cyclic aliphatic nitramine (Figure 1.1).

It is considered the most powerful of the military high explosives and, as such, is widely

used. During the late 1970s, RDX production was second only to TNT among the high

explosives (Ryon et al., 1984). Developed during World War II, RDX is synthesized

through either the direct nitration of hexamine using the Woolwich process or through

indirect nitration using the Bachmann process (Yinon, 1990). The solubility of RDX is

less than that of TNT (~60 mg/L), which has helped minimize its spread through aquatic

systems.

O2N NO2

2,4,6 trinitrotoluene

hexahydro-1,3,5-trinitro-1,3,5-triazine

NO2O2N

Figure 1.1. Explosive nitroaromatic compounds of concern to the Department of

Defense. TNT and RDX are highly recalcitrant, toxic compounds that are widespread in

soil, as well as ground and surface water, at military installations.

TOXICITY

During the early stages of World War I, large-scale manufacture of TNT resulted

in thousands of cases of “TNT intoxication” and hundreds of deaths (Palmer et al., 1996).

With the onset of World War II, production was increased substantially. However, far

fewer TNT-related illnesses and deaths were reported because of an alteration in the

production process that removed a hazardous byproduct, tetranitromethane (TNM), from

the final product (Bucher, 1990). Despite this, TNT toxicity remains a significant

problem as exposure to even low doses can have severe long-term consequences due to

the compound’s mutagenic properties.

Low-level TNT intoxication in humans is characterized by gastrointestinal

disorders, sometimes with associated tightening of the chest. In cases of severe

intoxication, cyanosis (a bluish or purplish tinge to the skin and mucous membranes) is

produced as a result of methemoglobin formation. (Methemoglobin is a brownish-red,

non-functional form of hemoglobin that occurs when hemoglobin is oxidized either

during decomposition of the blood or by the action of various oxidizing drugs or toxic

agents). Later stage symptoms include liver damage, aplastic anemia, cataract formation,

menstrual disorders, and neurological disorders (Ermakov et al., 1969). As noted above,

TNT has been reported to be mutagenic, even in the absence of metabolic activation,

producing frame-shift reversal mutations in Salmonella typhimurium (Dilley et al., 1978;

Ellis et al., 1978b). Not surprisingly, TNT is a carcinogenic substance producing various

tumors in the bladders of rodents (Furedi et al., 1984a). Its toxicity to aquatic organisms,

including marine copepods (Tigriopus californicus), oysters (Crassostrea gigas), and

freshwater unicellular green algae (Selenastrum capricornutum), is well established (Won

et al., 1976), as toxicological studies using these organisms revealed that TNT

concentrations as low as 2.5 to 10 mg/L could inhibit physiological functions. As a result

of this, the recommended maximum drinking water concentration for TNT is 140 µg/L

(Stahl and Aust, 1995).

Numerous studies have confirmed adverse effects of RDX exposure on the

nervous system, liver, and kidneys, as well as induction of testicular degeneration and

inflammation of the prostate in mice and rats. Inhalation is known to induce nausea,

irritability, confusion, and convulsions in exposed humans (Kaplan et al., 1965; Ketel and

Huyhes, 1972). In addition, hypotriglyceridemia, behavioral changes, and mortality have

also been observed in F344 rats fed toxic doses of RDX (Levine et al., 1990a). Due to

this acute toxicity in rodents, RDX has often been employed as a rat poison. Although

not thought to be mutagenic, one study reported a dose-related increase in hepatocellular

carcinomas and adenomas in female mice (Lish et al., 1984) suggesting carcinogenic

potential of RDX. Moreover, hexahydro-1,3,5-trinitroso-1,3,5-triazine, a degradation

product of RDX has been used as an experimental tumorigen (Lewis, 1992). RDX has

also been found to be quite toxic to various aquatic species (Bentley et al., 1977).

Bluegill used in acute exposure tests, displayed an LC50 of 3.6 mg/L after 96 hours of

exposure. In the same report, chronic exposure studies using fathead minnows produced

LC50 values of 6.3 mg/L. The toxicity of RDX to plants has also been evaluated (Cataldo

et al., 1990). At soil concentrations of 50 mg/kg RDX, bush beans (Phaseolus vulgaris),

wheat (Triticum aestivim), and bromegrass (Bromus inermis) became chlorotic and

displayed reduced plant heights, indicating a toxic response. As a result of this

widespread toxicity, water concentrations of RDX above 105 µg/L are considered

unacceptable.

ENVIRONMENTAL CONTAMINATION BY EXPLOSIVES AND METHODS OF

REMOVAL

The widespread use of explosives by the military has led to extensive

contamination of soil, as well as surface and ground water at processing and

manufacturing installations throughout the United States and Europe. At least 50 sites

identified by the U.S. Army Environmental Center (USAEC) (Figure 1.2) are in need of

remediation (Jerger and Woodhull, 2000), and a large number of other contaminated sites

have been identified in the U.K., Canada, and Australia. Clean-up efforts are underway

at many sites in the U.S., but relatively little clean-up is in progress in other countries

where the problem has only recently been recognized. In Germany the situation has been

further complicated by the demolition of explosives manufacturing facilities after World

War II, followed by the reconstruction of industrial and residential complexes over these

sites. Very little characterization has been done on these sites, and only a few projects

aimed at cleaning up these locations have been initiated.

Contamination became widespread at manufacturing and processing facilities due

to the rush to produce these explosives during the World Wars. In many instances, the

explosives leached from disposal lagoons into the surrounding soil. In other cases, TNT-

laden wastewater or “pink water”, so named for the color TNT imparts to aqueous

solutions, was released into surface waters, thereby spreading the contamination.

ties i

Obviously, removal of these compounds from the environment is of high priority

to the Department of Defense. Methods for removal have been developed, but all possess

undesirable characteristics and are in need of improvement. Typically, soil at highly

contaminated sites is treated through excavation followed by incineration or composting.

These strategies are not only laborious and costly, but concerns regarding toxicity and

mutagenicity of the transformation products resulting from these processes remain (Tan

et al., 1992; Jarvis et al., 1998). In addition, incineration produces unusable ash and has

been met with public opposition due to concern over air emissions. Estimated costs for

using incineration range between $200 and $1000 per cubic yard (Vanderford, 1996). On

top of this, excavation of contaminated soils is an ecologically destructive practice,

leaving sites heavily disturbed and adversely impacting nearby ecosystems.

Water polluted by the presence of explosives can be effectively treated using

carbon adsorption columns, but the spent carbon cannot be safely regenerated. This

prevents reuse of these materials and consequently drives up the cost of treatment. Due

to the inherent destructiveness and/or costs associated with these current methods,

researchers are seeking less intrusive, more environmentally friendly strategies to remove

TNT and RDX from the environment.

BIOREMEDIATION

An effective and potentially less destructive, as well as less costly, alternative to

excavation is bioremediation--the use of organisms to remove or degrade pollutants from

contaminated sites. In terms of costs, bioremediation offers a significant reduction for

clean-up projects. It has been estimated that microbial bioremediation in general (not

explosives exclusively) can cost as little as $25/cubic yard, compared to methods, such as

pump/treat or soil excavation, which typically cost $50-$100/cubic yard (Watanabe,

2001). Despite being a relatively new technology, many organisms have been identified

that possess abilities for contaminant degradation and removal. These include numerous

species of bacteria, certain fungi, and a number of plants.

BIOREMEDIATION OF TNT AND RDX

Bacteria

A large number of microbes have been found capable of metabolizing explosives

and both aerobic and anaerobic degradation mechanisms have been observed. Aerobic

metabolism of TNT often occurs through the removal of nitro groups from the toluene

ring following the formation of hydroxylamine aromatic compounds (e.g., 2-

hydroxylamino-4,6-dinitrotoluene). Alternatively, the formation of a Meisenheimer

complex, in which the ring is reduced via addition of a hydride ion (contributed by

reduced pyridine nucleotides) followed by nitro group loss (Esteve-nunez et al., 2001),

can also lead to nitro group removal. In general, aerobic bacteria have difficulty

metabolizing TNT to a significant extent, and few studies have reported the complete

mineralization of TNT in aerobic microbial cultures (Traxler et al., 1974; Bae et al.,

1995). Only a handful of studies have reported the use of TNT as a carbon or nitrogen

source by bacteria (Duque et al., 1993; Montpas et al., 1997; Oh and Kim, 1998).

Anaerobic metabolism of TNT by prokaryotes involves the complete reduction of

nitro groups to form 2,4,6-triaminotoluene (TAT), typically using H2 as an electron

donor. TAT formation in aerobic systems is highly unlikely due to the requirement for a

reduction potential around –200 mV. Although it is widely believed that TAT is further

metabolized through reductive elimination of the amino groups (Schnell and Schinck,

1991) there are no reports of deaminating reactions or toluene formation in anaerobic

cultures supplemented with TNT. However, strains of Desulfovibrio sp. have been

observed to grow on TNT as the sole nitrogen source (Boopathy and Kulpa, 1992; Preuss

et al., 1993).

Whereas extensive research has been conducted on the metabolism of TNT in

bacterial systems, very little information has been obtained for RDX. However, it

appears as though some microbes are capable of effectively degrading this explosive.

Under anaerobic conditions, RDX was metabolized through successive reduction of the

nitro substituents to hydroxyl-amino derivatives (the formation of hydrazine and

dimethylhydrazine, both known mutagens, was observed in the culture medium)

(McCormick et al., 1981). RDX degradation was also observed in liquid cultures of

Desulfovibrio sp. in which ammonia derived from the compound’s nitro groups served as

the sole nitrogen source (Boopathy et al., 1998a). Although there has been much

speculation on the metabolic mechanisms responsible for the degradation of RDX by

bacteria, virtually nothing has been confirmed.

The discovery that certain fungi have the ability to mineralize TNT has made this

class of organisms a major contender for the remediation of sites contaminated with TNT,

RDX, and other explosives. The first observation that fungi possess this ability occurred

in a study of Phaenerochaete chrysosporium grown under nitrogen-limiting conditions

(Hawari et al., 1999). Subsequent work showed that TNT may be mineralized under a

variety of nutrient starvation conditions (carbon, nitrogen, or sulfur limitation), and that

the reactions were carried out by enzymes secreted as part of the lignin-degrading system

of this white rot fungus (Stahl and Aust, 1995). As was seen in studies with bacteria,

reduction of the TNT nitro groups was the initial step in fungal degradation of this

compound, and in P. chrysosporium, this step appears to occur through the activity of a

membrane-bound nitroreductase that requires NADPH as a cosubstrate (Rieble et al.,

1994). The reaction is sensitive to molecular oxygen. Degradation continues under

lignolytic conditions, ultimately resulting in complete mineralization of TNT, most likely

through the activity of lignin peroxidases (Michels and Gottschalk, 1995; Hodgson et al.,

2000). Despite this ability to mineralize TNT, P. chrysosporium tolerates relatively low

levels (<20 mg/kg) of the explosive, making field application of this organism unrealistic.

Other fungal species display a greater level of tolerance to TNT (e.g., Cladosporium

resinae), but do not appear capable of mineralizing the compound (Bayman and Radkar,

1997).

Little is known of the potential for fungi to degrade RDX. One study reported the

mineralization of RDX under aerobic conditions, although the mechanism of degradation

was not described (Fernando and Aust, 1991).

Plants

Since the mid-1990s, phytoremediation, or the use of plants for the restoration of

contaminated soils, groundwater, or surface water, has been recognized as a potentially

effective technology (Burken, 2000). Although a relatively new technology,

phytoremediation is already showing promise for the clean up of a wide variety of

pollutants and may well prove effective for the removal of explosives from contaminated

soil and water. Not only is phytoremediation an efficient method for contaminant

removal, it has the advantage of being environmentally friendly and offers substantial

financial savings. For example, where excavation and incineration may cost $200-

$1,500/ton, and in situ bioremediation may cost around $50-$100/ton, phytoremediation

may cost as little as $10/ton of soil (Watanabe, 2001). The promise of this technology is

significant and has stimulated extensive research into its potential for addressing a

number of the most threatening and widespread pollutants.

PHYTOREMEDIATION

Phytoremediation has become an attractive alternative for environmental cleanup

for several reasons: 1) plants quickly stabilize surface soil, limiting the runoff of

contaminants into nearby streams; 2) plant roots can penetrate soil to reach and extract

contaminants buried in the subsoil; 3) plants provide their own energy for the requisite

degradative processes; 4) plant root exudates can stimulate the growth of soil microbes

that further contribute to the breakdown of organic contaminants; and 5) plants have the

potential to be deployed on disturbed sites in a cost-effective manner that minimizes

further disruption of the surface soil and consequent spreading of the contaminant (e.g.,

aerial seeding).

The number of contaminant attenuation mechanisms possessed by plants makes

their use in remediating contaminated land and water feasible. As a result of their

sedentary nature, plants have evolved diverse abilities for dealing with toxic comounds in

their environment. It is thought that plants have the ability to “synthesize, rearrange, and

detoxify the most complex array of biochemicals and biopolymers of any living

organisms” (Meagher, 2000). Furthermore, “plants control most of the energy in an

ecosystem, and usually account for several orders of magnitude greater biomass than any

few bacterial species in the soil” (Meagher, 2000).

Indirect mechanisms involving plant-mediated modifications to the immediate

surrounding environment add to these abilities. It is well documented that plant-induced

changes to the contaminated environment (alteration of redox conditions, shifts in

microbial community composition, and modificiation of site hydrodynamics) serve as an

indirect means of remediation. For example, plants may secrete 10-20% of their

photosynthate in root exudates to stimulate the growth of diverse bacterial and fungal

communities in the rhizosphere (Lynch, 1982; Haselwandter, 1983; Shimp et al., 1993;

Anderson et al., 1994). Populations of these organisms are significantly greater (two to

four orders of magnitude) in vegetated versus non-vegetated soils. In addition, these

specific rhizosphere-associated organisms typically possess a wide range of metabolic

capabilities, including those required for the degradation of recalcitrant xenobiotic

compounds (Anderson et al., 1994; Walton et al., 1994). Typical root exudates include

phenolic compounds, organic acids, alcohols, and various proteins that serve as carbon

and nitrogen sources for the bacteria and fungi. These plant products provide not only for

the growth and long-term survival of soil microbes, but also serve as a supply of co-

metabolites, such as catechin and coumarin, that facilitate microbial breakdown of

compounds like poly-chlorinated biphenyls (PCBs) (Bedard et al., 1987; Donnelly et al.,

1994; Fletcher and Hedge, 1995; Hedge and Fletcher, 1996). It is for this reason that

phytoremediation technology is often described as “plant-assisted” remediation (Burken

et al., 2000)

With respect to their direct roles in remediation processes, plants use several

different strategies for dealing with environmental chemicals: phytoextraction,

phytodegradation, phytovolatilization, and rhizodegradation (Schnoor, 1997).

Phytoextraction involves the removal and subsequent storage of contaminants by the

plant. This term is frequently applied to the removal and storage of heavy metals that

may undergo electrochemical transformation in plants, but cannot be degraded.

However, certain organic chemicals may also be treated in this manner due to inherent

resistance to degradation. Conversely, phytodegradation describes processes in which

plants metabolize the contaminants they take up. Components of this mechanism are

often utilized by plants exposed to herbicides and thus have been researched extensively.

The metabolic processes involved in phytodegradation have strong similarities to those

used by animals for modification and degradation of drugs and other toxins. This has

given rise to a conceptual model for phytodegradation known as the “green liver” model,

which will be discussed below later in more detail (Sandermann, 1994). A further

attenuation mechanism, referred to as phytovolatilization, involves the release of

contaminants to the atmosphere following their uptake from the soil or water. This

mechanism has been observed for both organic and heavy metal contaminants, including

trichloroethylene (TCE), which has been observed in the off-gas from plant leaves in the

laboratory and field (Compton et al., 1998), and in the production of volatile, elemental

mercury by genetically-engineered Arabidopsis thaliana grown in the presence of ionic

mercury (Rugh et al., 1996; Bizily et al., 1999).

An indirect mechanism, rhizodegradation refers to the transformation of

contaminants by resident microbes in the plant rhizosphere (i.e., the microbe-rich zone in

intimate contact with the root vascular system). As mentioned above, the presence of

plants on contaminated sites can drastically affect soil redox conditions and organic

content (often through the secretion of organic acids from roots), as well as soil moisture.

Rhizodegradation is the dominant mechanism in the removal of total petroleum

hydrocarbons from soil by deep-rooted trees (Carman et al., 1998), as well as annual

species (Schwab and Banks, 1994).

PLANT METABOLISM OF TNT AND RDX

Numerous studies have been conducted to understand the fate of TNT in

terrestrial and aquatic plants (Dacre and Rosenblatt, 1974; Smock et al., 1976; Palazzo

and Leggett, 1986; Cataldo et al., 1990; Harvey et al., 1991; Gorge et al., 1994; Wolfe et

al., 1994; Medina et al., 1996; Medina, 1996; Best et al., 1997; Larson, 1997; Vanderford

et al., 1997; Lauritzen, 1998; Scheidemann et al., 1998; Thompson et al., 1998; Best et

al., 1999; Bhadra et al., 1999a, b; Larson et al., 1999b; Larson et al., 1999a; Wayment et

al., 1999). TNT was never recovered stoichiometrically in these studies, suggesting the

existence of metabolic pathways for its transformation and possible degradation. In

contrast, RDX typically accumulates in plant tissues with little evidence of degradation.

TNT Metabolism

Although the metabolism of TNT in microbial systems has been studied

extensively, until recently little was known about its metabolism in plants. Studies of

TNT metabolite mass balances confirmed the existence of plant mechanisms for the

uptake and transformation of this explosive (Hughes et al., 1997). Researchers employed

uniformly labeled 14C-TNT to determine the fate of TNT in Catharanthus roseus hairy

root cultures, axenic Myriophyllum aquaticum plants, and wild-grown Myriophyllum

aquaticum. In all three systems, the TNT was completely transformed, although

mineralization was not observed. The aminodinitrotoluenes (2-amino-4,6-dinitrotoluene

and 4-amino-2,6-dinitrotoluene) were observed in the growth media, as had been

observed in microbial studies. However, unidentified 14C-labeled soluble and extractable

products were observed, but could not be identified, as reduction products (i.e.,

aminonitrotoluenes, hydroxyl-aminonitrotoluenes, or azoxynitrotoluenes). In addition,

radiolabel also ended up in bound residues (i.e., a non-extractable plant-associated

material that could only be quantified after combustion of the plant tissue). The soluble

and solvent-extractable products comprised the majority of TNT derivatives. The

production of unidentifiable soluble products hinted at the involvement of plant

conjugation reactions. This notion was later confirmed through the identification of

conjugates in which various six-carbon sugars were bound to the TNT transformation

products (Bhadra et al., 1999a).

The identification of various TNT transformation products and conjugates lead to

conceptualization of the “green liver” model for plant metabolism of explosives

(Sandermann, 1994). This model was born out of the idea that plants use

“detoxification” processes similar to those observed in the metabolism of xenobiotics by

animals. This is in contrast to microbial heterotrophic metabolism in which foreign

compounds are typically broken down for energy via the diverse catabolic pathways

available to microbes. Being autotrophic, plants do not possess this range of catabolic

pathways and consequently tend to detoxify xenobiotics rather then utilizing them as

carbon or nitrogen sources. It is through the multi-phase mechanism proposed by the

green liver model that this detoxification is often achieved.

Studies of plant metabolism of herbicides and other xenobiotics first revealed the

existence of this multi-phase detoxification mechanism (Klein and Scheunert, 1982). In

the initial phase of detoxification, following entrance of the compound into the cell,

transformation reactions introduce chemical groups amenable to subsequent conjugation

reactions. This is frequently achieved through either enzyme-catalyzed oxidation

reactions involving cytochrome P450 enzymes and peroxidases (Bhadra et al., 1999b) or

through enymze-catalyzed reduction or hydrolysis reactions. Once transformed, the

metabolized derivative may be conjugated to glutathione (GSH) or six-carbon sugars by

glutathione S-transferases (GSTs) or UDP:glucosyltransferases, respectively. This

constitutes the second phase of detoxification. Introduced moieties (OH-, NH2-, SH-, and

COOH) on transformed molecules are typically conjugated to sugars (Frear, 1976), while

conjugated double bonds, halogen- or nitrofunctions typically promote conjugation to

glutathione. In the third phase of detoxification, conjugates are sequestered into plant

storage organelles and/or released to the extracellular space. At this point, the conjugate

may be either immobilized through the addition of further substitutions or degraded. In

most cases, the conjugates are rapidly removed from the cell, likely to prevent the

inhibition of conjugating enzymes as it has been found that glutathione conjugates inhibit

both GST and glutathione reductase (Schröder and Wolfe, unpublished). Transportation

of conjugates is often achieved through the action of ABC-type ATPases that appear to

recognize the sugar or glutathione groups, rather than the xenobiotic compound itself

(Schroder and Collins, 2002).

RDX Metabolism

This multi-phase mechanism for TNT detoxification in plants has not been

observed for metabolism of RDX. Although few studies have examined the fate of RDX

in plants, it appears that the compound is much more resistant to transformation than

TNT. Using Catharanthus roseus and Myriophyllum aquaticum, the disappearance of

RDX occurred at a much slower rate than for TNT (Bhadra et al., 2000). Mass balance

studies in C. roseus showed that the majority of 14C-labeled RDX remained

untransformed after entering the plant (Bhadra et al., 2000). This phenomena was

previously noted by Harvey et al. (1991), in a study of bush beans grown for several days

in a hydroponic solution containing 10 ppm RDX. The plants took up 60% of the

available RDX in this time, but the compound remained virtually untransformed in the

plant (Harvey et al., 1991). However, after 60 days, a fraction (30%) of the RDX

appeared to have undergone some sort of transformation as various polar derivatives were

detected in root tissues, and 50% of the RDX label was no longer extractable. Analysis

of foliar tissues revealed that RDX concentrations were in excess of 90 mg/kg, suggesting

a mechanism for phytoextraction of this explosive. Similar results were seen in hybrid

poplars where a large percentage of 14C-radiolabeled RDX was identified in the leaves

(Thompson et al., 1999). Accumulation of RDX in the aboveground portions of

terrestrial plants growing in RDX-contaminated soil is of some concern for

phytoremediation processes. Ingestion of these tissues by insects and herbivores could

introduce the compound into foodchains, thereby reducing the benefits of a

phytoremediation strategy for cleanup. The use of plants to remediate RDX-

contaminated soil may also be limited by the inability of plants to remove RDX

efficiently from soils having high organic content. It has been observed that while uptake

rates are relatively high in soils having low organic content, uptake is drastically reduced

in the same soils when organic matter is added (Chen, 1993).

While many plants have innate abilities to remove explosives and other organics

from aqueous solutions (Best, 1997, 1999; Bhadra, 1999), few are robust enough to

survive and remediate the high levels of contamination common to most of the sites

known today. In an attempt to address this situation, researchers have designed

transgenic plants with enhanced abilities to tolerate and remove TNT from their

environment (French, 1999; Hannink, 2001). However, current understanding of the

biochemical mechanisms involved in the processes of uptake and degradation of TNT

and RDX in plants is rudimentary and limits our ability to manipulate metabolism

efficiently. Studies of plant-explosives interactions have been limited to toxicity

evaluations and identification of metabolites, which allows only for inference of the

enzymes that might be involved. If phytoremediation is to become an effective

technology for the removal and degradation of explosives, a more refined knowledge of

the mechanisms used by plants for these processes must be obtained. Elucidation of

these mechanisms would facilitate engineering of plants designed for not only removal

and degradation of explosives, but also for tolerance and persistence in environments

where contaminant levels are too high to permit survival of wild-type plants. One

promising strategy for identifying the pertinent metabolic pathways involves use of

genomic techniques to determine changes in gene expression induced by contaminant

exposure. The induction of specific plant genes in response to explosives should

highlight specific enzymes responsible for conferring tolerance or degradative abilities to

plants.

GENOMICS

The characterization of complete genome sequences for a number of organisms,

and the emergence of new technologies has created a situation where the expression of

large numbers of genes can be studied simultaneously under any chosen condition. For

many organisms (e.g., S. cerevisiae) total genome expression analysis is now routine

(Lashkari et al., 1997; Wodicka et al., 1997), and for many higher eukaryotes, expression

analyses covering a significant portion of the genome are possible. The massive amounts

of data generated by these analyses can present difficult bioinformatics challenges, but

the information that can be gleaned from such studies provides new insights into cellular

and genome function (Meltzer, 2001). In addition to speeding discoveries of fundamental

biological functions, such studies are leading to significant advances in our understanding

of human disease, particularly cancers (Khan et al., 1998; Alon et al., 1999; Golub et al.,

1999; Alizadeh et al., 2000; Bittner et al., 2000; Ross et al., 2000; Scherf et al., 2000).

Currently, the most commonly used genomic techniques are aimed at measuring changes

in the transcriptome. As transcription represents the initial expression of the genome,

comparisons of transcriptome profiles for different tissues (e.g., treatment versus control)

can be a powerful approach to the study of metabolism and cellular function.

TRANSCRIPTIONAL PROFILING

A comparison of transcript abundance between different tissues or treatments has

proven to be an effective method for identifying differentially expressed genes. While it

must be recognized that transcript levels do not always reflect the levels of corresponding

proteins in the cell, and that post-translational modifications are common (Gygi et al.,

1999), it is generally accepted that in the modulation of gene expression, the primary

level of control resides at the level of transcription (Donson et al., 2002). Based on this

assumption, a number of technologies have been developed to allow for the rapid

quantification of different gene transcripts simultaneously. Current methods are based on

direct analysis of the transcriptome using DNA sequencing or fragment sizing, or indirect

analysis using nucleic acid hybridization.

Methods for Direct Analysis of the Transcriptome

Nucleotide Sequencing-based methods

Methods of transcriptional profiling that rely on DNA sequencing include large-

scale expressed sequence tag (EST) sequencing, serial analysis of gene expression

(SAGE), and massively parallel signature sequencing (MPSS).

Large-scale Expressed Sequence Tag (EST) Sequencing

EST sequencing projects provide both the potential to discover previously

unidentified genes and the ability to assess differential gene expression. By measuring

the frequency of occurrence of a particular EST in each cDNA library, the induction or

repression of that gene can be determined in relation to other libraries. This approach

also provides long sequences that facilitate identification of gene function. Thus, this

technology is not as dependent as other techniques on established sequence data for the

organism under study. However, the large amount of DNA sequencing required makes

this technology costly and its routine use for profiling unrealistic at present.

Furthermore, size and sequence composition biases in cloning and cDNA synthesis limit

the comprehensiveness of this approach (although these limitations are not unique to EST

technology).

Serial Analysis of Gene Expression (SAGE)

Issues of sequencing costs for in-depth analyses of EST libraries was addressed in

part through development of the SAGE technology (Velculescu et al., 1995). Like EST

sequencing, SAGE is an open-ended technique that allows for the analysis of all

transcripts in a given sample and thus is an excellent tool for gene discovery. In brief,

SAGE generates short (14-mer) cDNA fragments known as tags from the 3’ end of

transcripts. Although short in sequence, these tags usually provide enough information to

identify the original transcript expressed in the tissue under study (Figure 1.3). The tags

are then linked together in long chains called concatemers, which are subsequently

cloned and sequenced. Using custom software, the tags can then be identified and

counted to determine the number of times a specific tag appears within all of the

sequences from a particular tissue sample.

One of the major drawbacks of the SAGE technology is the short length of the

tags. This characteristic can sometimes make unambiguous gene identification difficult,

although the potential for misidentification is drastically reduced by the fact that the tags

are generated from a specific location within each transcript (the NlaIII restriction site

closest to the 3’ end of the transcript). SAGE is also most effective where a sequenced

genome is available for tag identification. In addition, as SAGE measures the abundance

of individual tags in a large pool of tags, the extent of sequencing can have a profound

effect on the validity of the data based on sampling effects. For example, an analysis of

1,000 tags will be much less reliable than an analysis of 20,000 tags (which can cost

considerably more).

Figure 1.2. General diagram of the SAGE concept showing the linkage of tags

together to form concatemers followed by sequencing and computational analysis.

Diagram from the sagenet website:

(http://www.sagenet.org/home/Description.htm)

Massively Parallel Signature Sequencing (MPSS)

Massively Parallel Signature Sequencing or MPSS provides a solution to both the

time required for depth of analysis in EST sequencing projects and the lack of sequence

data included in SAGE tags (Brenner et al., 2000a; Brenner et al., 2000b). The technique

is based on use of microbeads for in vitro cloning of cDNA fragments generated from an

mRNA population. This is achieved by synthesizing random oligonucleotides (32-mers)

or “tags” from a defined set of 4-mers, and then attaching them to the generated cDNA

fragments such that each fragment possesses a unique tag. These cDNA-tag conjugates

are then amplified using PCR and subsequently hybridized to complementary sequences

(anti-tags) attached to microbeads. As each microbead possess approximately 100,000

copies of a single anti-tag sequence, roughly the same number of cDNA-tag conjugates

can be bound to one microbead. The cDNA fragments are then sequenced in a flow cell,

and the number of beads displaying the same cDNA fragment are counted and summed

for comparison. Typically sequences of 16-20 bases from each cDNA fragment are

recorded, thereby increasing the amount of sequence data derived from each transcript

over that obtained using the SAGE technique (Donson et al., 2002). As the procedure

can be completed in a matter of a few days, MPSS combines both the digital nature of

EST sequencing projects with a vastly superior throughput. In addition, the large number

of sequence tags that can be measured (1,619,000 in a single study of human acute

monocytic leukemia cells (Brenner et al., 2000a)) provides for more robust statistical

analyses, producing more reliable data. The major drawback to this technology is its

proprietary nature. The technology is both owned and executed by Lynx Therapeutics,

Inc., and the cost of access keeps it out of reach for most academic researchers.

Fragment sizing-based methods

Methods of direct analysis that rely on DNA fragment sizing include differential

display (DD) and cDNA-AFLP (and variations of these). These methods utilize physical

separation of cDNA fragments representing mRNAs through gel electrophoresis and

selective PCR amplification.

Differential Display (DD)

The first application of fragment-sizing based methods for transcriptional

profiling was differential display (DD) (Liang and Pardee, 1992). Differential display

requires the reverse transcription of mRNA primed by anchored oligo dT primers

containing a specific nucleotide at the 3’ ends. These cDNA fragments are then

amplified by PCR using the anchored primers and an oligonucleotide primer of arbitrary

sequence. The PCR products are separated on a polyacrylamide gel and visualized using

autoradiography (radioactive dATP is supplied during PCR amplification to allow for

product visualization). Comparing the bands for mRNA populations originating from

two different cell populations (i.e., different tissues, treatments, etc.), differences in

banding patterns reveal differential expression of specific transcripts (Zhang et al., 1998).

Although a relatively cost-effective method for the analysis of differential gene

expression, differential display possesses shortcomings that deter its continued use given

alternative technologies. In particular, the PCR strategy approach attempts to amplify

specific cDNAs using arbitrary primers at low annealing temperatures so as to allow for

priming at multiple sites. Thus, amplification of cDNAs is based not only on their initial

concentrations, but also on the quality of match between the primers and the template

(Matz and Lukyanov, 1998). This situation results in detection of numerous false

positive bands after electrophoresis (Sun et al., 1994; Sompayrac et al., 1995) and limits

reproducibility (Haag and Raman, 1994; Zhang et al., 1998). In addition, the technique

does not quantify the degree of differential expression, although its sensitivity does allow

for detection of changes in relatively rare transcripts.

cDNA-AFLP

A modification of the original differential display protocol, cDNA-AFLP

(amplified fragment-length polymorphism) attempts to counteract the problems

mentioned above through the use of more stringent PCR conditions provided by the

ligation of adaptors to restriction fragments, and the use of specific primer sets (Vos et

al., 1995; Bachem et al., 1996). Unlike differential display, cDNA-AFLP (and other

similar techniques) allows for the systematic survey of an organism’s transcriptome

through the use of selective fragment amplification. cDNA-AFLP has been shown to

produce both good reproducibility and sensitivity (Donson et al., 2002). However, like

differential display, the technique still requires extensive band isolation and DNA

sequencing, making it less suitable to high-throughput analyses than other transcriptional

profiling approaches.

Methods for Indirect Analysis of the Transcriptome

Methods for indirect analysis of the transcriptome rely on the hybridization of

complementary sequences. While this characteristic has been exploited for decades,

(Gillespie and Spiegelman, 1965) until fairly recently this has been on only a relatively

small scale. Due to the recent explosion in the amount of sequence data and routine

methods for cloning, hybridization-based approaches now provide a means to study the

expression of thousands of genes simultaneously. These approaches most often employ

DNA microarray technology. DNA microarrays consist of specific DNA elements

(“probes”) affixed to a solid support to which labeled cDNAs (“targets”) derived from

cellular mRNAs expressed under the test condition(s) may be hybridized (Figure 1.4).

The probes may be either cDNAs amplified by PCR from cloned sequences or

synthesized oligonucleotides representing unique gene sequences. Unlike previously

described technologies, DNA microarrays only allow expression measurement for genes

corresponding to those fabricated into the array. However, high-density microarrays now

commonly in use may contain more than 400,000 elements, and the fabrication of 60

million elements on a single array is now possible (Perlegen Sciences, Mountain View,

CA). This provides the promise of routine analysis of the genetic expression of entire

higher eukaryote genomes in the near future.

Microarrays, although close-ended and, therefore, less comprehensive, offer some

advantages over the other technologies mentioned. Foremost, microarrays allow for fast

analysis, as the spotting of genes or oligos is a rapid, automated process, and subsequent

hybridization and scanning require little time. Thus, microarrays can be used to quickly

profile transcripts across numerous conditions, which would be prohibitively time

consuming and costly using other techniques. Although not as vulnerable to sampling

effects as some other techniques (e.g., SAGE and EST sequencing projects), microarrays

are affected by hybridization efficiencies that can vary considerably across array

elements (Meltzer, 2001). Finally, microaray technology is relatively affordable and

open to the general scientific community, making its widespread use a reality.

PHYTOREMEDIATION AND GENOMICS

In the case of plant responses to explosives, virtually nothing is known with

regard to the genes that respond to these toxic compounds. Thus, both SAGE and

microarrays have great potential to contribute to our understanding of the processes

plants use in dealing with these compounds. As the plant model organism, Arabidopsis is

ideally suited to these types of studies. With a sequenced genome (The Arabidopsis

Genome Initiative, 2000), extensive bioinformatic resources, ease of growth and

handling, established methods for transformation, and a wealth of biochemical and

molecular understanding in the scientific community, Arabidopsis provides a means for

taking full advantage of the strength of transcript profiling technologies. Genomic

studies of Arabidopsis responses to explosives will accelerate the discovery of genes of

potential use in the engineering of plants with enhanced abilities for explosives tolerance,

removal, and degradation.

Figure 1.4. Simplified overview of the method for sample preparation and hybridization

to cDNA microarrays. For illustrative purposes, samples derived from cell culture are

depicted, although other sample types are amenable to this analysis (Nuwaysir et al.,

1999).

REFERENCES

Alizadeh, A.A., Eisen, M.B., Davis, R.E., Ma, C., Lossos, I.S., Rosenwald, A.,

Boldrick, J.C., Sabet, H., Tran, T., Yu, X., Powell, J.I., Yang, L., Marti, G.E.,

Moore, T., Hudson, J., Jr., Lu, L., Lewis, D.B., Tibshirani, R., Sherlock, G.,

Chan, W.C., Greiner, T.C., Weisenburger, D.D., Armitage, J.O., Warnke, R.,

Staudt, L.M., et al. (2000). Distinct types of diffuse large B-cell lymphoma

identified by gene expression profiling. Nature 403, 503-511.

Alon, U., Barkaai, N., Notterman, D.A., Gish, K., Ybarra, S., Mack, D., and Levine,

A.J. (1999). Broad patterns of gene expression revealed by clustering analysis of

tumor and normal colon tissues probed by oligonucleotide arrays. Proc. Natl.

Acad. Sci. USA 966, 6745-6750.

Anderson, T.A., Kruger, E.L., and Coats, J.R. (1994). Enhanced degradation of a

mixture of three herbicides in the rhizosphere of a herbicide-tolerant plant.

Chemosphere 28, 1551-1557.

Bachem, C.W.B., van der Hoeven, R.S., de Brujin, S.M., Vreugdenhil, D., Zabeau,

M., and Visser, R.G.F. (1996). Visualization of differential gene expression

using a novel method of RNA fingerprinting based on AFLP: Analysis of gene

expression during potato tuber development. Plant J. 9, 745-753.

Bae, B., Autenrieth, R.L., and Bonner, J.S. (1995). Aerobic biotransformation and

mineralization of 2,4,6-trinitrotoluene. In Bioremediation of recalcitrant organics,

R.E. Hinchee, Hoeppel, R.E., and Anderson, D.B., ed (Columbus, OH: Battelle

Press), pp. 231-238.

Bayman, P., and Radkar, G.V. (1997). Transformation and tolerance of TNT (2,4,6-

trinitrotoluene) by fungi. Int. Biodeterior. Biodegrad. 39, 45-53.

Bedard, D.L., Wagner, R.E., Brennen, M.J., Haberl, M.L., and Brown, J.F.J. (1987).

Extensive degradation of aroclors and environmentally transformed

polychlorinated biphenyls by Alcaligenes eutrophus H850. Appl. Environ.

Microbiol. 53, 1094-1102.

Bentley, R.E., Dean, J.W., Ellis, S.J., Hollister, T.A., LeBlanc, G.A., Sauter, S., and

Sleight III, B.H. (1977). Laboratory Evaluation of the Toxicity of RDX to

Aquatic Organisms. Contract DAMD17-74-C-4101. EG&G Bionomics,

Wareham, MA. DTIC AD-A061730.

Best, E.H.P., Miller, J.L., Fredrickson, H.L., Larson, S.L., Zappi, M.E., and

Streckfuss, T.H. (1997). Screening of aquatic and wetland plant species for the

phytoremediation of explosives-contaminated groundwater from the Iowa Army

Ammunition Plant. Ann. N.Y. Acad. Sci. 829, 179-194.

Best, E.H.P., Sprecher, S.L., Larson, S.L., Fredrickson, H.L., Bader, D.F. (1999).

Environmental behavior and fate of explosives from groundwater from the milan

army ammunition plant in aquatic and wetland plant treatments. Chemosphere 38,

3383-3396.

Best, E.H.P., Zappi, M.E., Fredrickson, H.L., Sprecher, S.L., Larson, S.L., Ochman,

M. (1997). Screeing of aquatic and wetland plant species for the

Ammunition. Plant. Ann. N.Y. Acad. Sci. 829, 179-194.

Bhadra, R., Wayment, D.G., Hughes, J.B., and Shanks, J.V. (1999a). Confirmation of

conjugation processes during TNT metabolism by axenic plant roots. Environ.

Sci. Technol. 33, 446-452.

Bhadra, R., Wayment, D.G., Hughes, J.B., and Shanks, J.V. (1999b). Characterization

of oxidation products of TNT metabolism in aquatic phytoremediation systems of

Myriophyllum aquaticum. Environ. Sci. Technol. 33, 3354-3361.

Bhadra, R., Williams, R., Barman, S., Stone, M.B., Hughes, J.B., and Shanks, J.V.

(2000). Fate of RDX and HMX in axenic plant roots. Chemosphere 44, 1259-

Bittner, M., Meltzer, P., Chen, Y., Jiang, Y., Seftor, E., Hendrix, M., Radmacher,

M., Simon, R., Ben-Dor, A., Sampas, N., Dougherty, E., Wang, E., Marincola,

F., Gooden, C., Lueders, C., Glatfelter, A., Pollock, P., Carpten, J.,

Gillanders, E., Leja, D., Dietrich, K., Beaudry, C., Berens, M., Alberts, D.,

Sondak, V., Hayward, N., and Trent, J.M. (2000). Molecular classification of

cutaneous malignant melanoma by gene expression profiling. Nature 406, 536-

Bizily, S., Rugh, C.L., Summer, A.O., and Meagher, R.B. (1999). Phytoremediation of

methylmercury pollution: merB expression in Arabidopsis thaliana confers

resistance to organomercurials. Proc. Natl. Acad. Sci. USA 96, 6808-6813.

Boopathy, R., and Kulpa, C.F. (1992). Trinitrotoluene as a sole nitrogen source for a

sulfate-reducing bacterium Desulfovibrio sp. (strain B) isolated from an anaerobic

digester. Curr. Microbiol. 25, 235-241.

Boopathy, R., Gurgas, M., Ullian, J., and Manning, J. (1998a). Metabolism of

explosive compounds by sulfate-reducing bacteria. Curr. Microbiol. 37, 127-131.

Brenner, S., Williams, S.R., Vermass, E.H., Storck, T., Moon, K., McCollum, C.,

Mao, J.I., Luo, S., Kirchner, J.J., Eletr, S., DuBridge, R.B., Burcham, T., and

Albrecht, G. (2000b). In vitro cloning of complex mixtures of DNA on

microbeads: physical separation of differentially expressed cDNAs. Proc. Natl.

Acad. Sci. USA 97, 1665-1670.

Brenner, S., Johnson, M., Bridgham, J., Golda, G., Lloyd, D.H., Johnson, D., Luo,

S., McCurdy, S., Foy, M., Ewan, M., Roth, R., George, D., Eletr, S., Albrecht,

G., Vermaas, E., Williams, S.R., Moon, K., Burcham, T., Pallas, M.,

DuBridge, R.B., Kirchner, J., Fearon, K., Mao, J.I., and Corcoran, K.

(2000a). Gene expression analysis by massively parallel signature sequencing

(MPSS) on microbead arrays. Nat. Biotechnol. 18, 630-634.

Bucher, J. (1990). Toxicology and Carcinogenesis Studies of Tetranitromethane (CAS

No 509-14-8) in F344/N and B6C3F1 Mice (Inhalation Studies). Technical

Report Series No. 386. National Toxicology Program, U.S. Department of Health

and Human Services, Public Health Service, Research Triangle Park, N.C.

Burken, J.G., Shanks, J.V., and Thompson, P.L. (2000). Phytoremediation and Plant

Metabolism of Explosives and Nitroaromatic Compounds. In Biodegradation of

Nitroaromatic Compounds and Explosives, J.C. Spain, ed,pp. 239-275.

Carman, E., Crossman, T., and Gatliff, E. (1998). Phytoremediation of no. 2 fuel-oil

contaminated soil. Journal of Soil Contam. 7, 455-466.

Cataldo, D.A., Harvey, S.D., and Fellows, R.J. (1990). An Evaluation of the

Environmental Fate and Behavior of Munitions Material (TNT, RDX) in Soil and

Plant Systems: Environmental Fate and Behavior of RDX. Final Report, Army

Project Order 88PP8853, PNL-7529, Pacific Northwest Laboratory, Richland,

Chen, D. (1993). M.S. thesis. University of Illinois, Urbana-Champaign.

Compton, H.R., Harosi, D.M., Hirsch, S.R., and Wrobel, J.G. (1998). Pilot-scale use

of trees to address voc contamination. In Bioremediation and Phytoremediation,

Chlorinated and Recalcitrant Compounds, G.B. Wickramanayake and R.E.

Hinchee, eds (Columbus: Battelle Press).

Dacre, J., and Rosenblatt, D.H. (1974). Mammalian toxicology and toxicity to aquatic

organisms of four important types of waterborne munitions pollutants-an

extensive literature evaluation. AD-A778725. TR-7403. U.S. Army Medical

Bioengineering Research and Development Laboratory, Fort Detrick, Fredrick,

Dilley, J.V., Tyson, C.A., and Newell, G.W. (1978). Mammalian Toxicological

Evaluation of TNT Wastewaters. Volume 2: Acute and Subacute Mammalian

Toxicity of TNT and LAP Wastewaters. Final Report, Contract DAMD17-76-C-

6050. SRI International, Menlo Park, CA. DTIC AD-A080957.

Donnelly, P.K., Hegde, R.S., and Fletcher, J.S. (1994). Growth of PCB-degrading

bacteria on compounds from photosynthetic plants. Chemosphere 28, 981-988.

Donson, J., Fang, Y., Espiritu-Santo, G., Xing, W., Salazar, A., Miyamoto, S.,

Armendarez, V., and Volkmuth, W. (2002). Comprehensive gene expression

analysis by transcript profiling. Plant Mol. Biol. 48, 75-97.

Duque, E., Haidour, A., Godoy, F., and Ramos, J.L. (1993). Construction of a

Pseudomonas hybrid strain that mineralizes 2,4,6-trinitrotoluene. J. Bacteriol.

175, 2278-2283.

Ellis, H.V., Hodgson III, J.D., Hwang, S.W., Halfpap, L.M., Helton, D.V., Anderson,

B.S., VanGoethem, D.L., and Lee, C.C. (1978b). Mammalian Toxicity of

Munitions Compounds Phase I: Acute Oral Toxicity, Primary Skin and Eye

Irritation, Dermal Sensitization, Disposition and Metabolism, and Ames Tests of

Additional Compounds. Report 6, Contract DAMD17-74-C-4073. Midwest

Research Institute, Kansas City, MO. DTIC AD-A060333.

Ermakov, E.V., Aizenschtadt, V.S., and Bentsenostsev, B.B. (1969). Chronic

trinitrotoluene poisoning (clinical picture and pathogenesis of neurovegetative

disturbances). Sov. Med. 32, 119-122.

Esteve-nunez, A., Caballero, A., and Ramos, J.L. (2001). Biological degradation of

2,4,6-trinitrotoluene. Microbiol. and Mol. Biol. Rev. 65, 335-352.

Fernando, T., and Aust, S.D. (1991). Biodegradation of munition waste, TNT (2,4,6-

trinitrotoluene), and RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine) by

Phanerochaete chrysosporium. ACS Symposium Series. Ind. Eng. Chem. 486,

214-232.

Fernando, T., Bumpus, J.A., and Aust, S.D. (1990). Biodegradation of TNT (2,4,6-

trinitrotoluene) by Phanerochaete chrysosporium. Appl. and Environ. Microbiol.

56, 1667-1671.

Fletcher, J.S., and Hedge, R.S. (1995). Release of phenols by perrennial plant roots and

their potential importance in bioremediation. Chemosphere 31, 3009-3016.

Frear, D.S. (1976). Pesticide conjugates-Glycosides, In: Bound and Conjugates

Pesticide Residues. In ACS Symposium, D.D. Kaufman, G.G. Still, G.D. Paulson,

and S.K. Bandal, eds (Washington, DC: American Chemical Society), pp. 35-54.

French, C.E., Rosser S.J., Davies, G.J., Nicklin, S., and Bruce, N.C. (1999).

Biodegradation of explosives by transgenic plants expression pentaerythritol

tetranitrate reductase. Nature Biotechnology 17, 491-494.

Furedi, E.M., Levine, B.S., Gordon, D.E., Rac, V.S., and Lish, P.M. (1984a).

Determination of the Chronic Mammalian Toxicological Effects of TNT.

Twenty-four Month Chronic Toxicity/Carcinogenicity Study of Trinitrotoluene

(TNT) in the Fischer 344 Rat. Phase III, Vols. 1-4. Final Report, Contract

DAMD17-79-C-9120. IIT Research Institute, Chicago, IL. DTIC AD-A168637.

Gillespie, D., and Spiegelman, S. (1965). A quantitative assay for DNA-RNA hybrids

immobilized on a membrane. J. Mol. Biol. 3, 829-842.

Golub, T.R., Slonim, D.K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J.P.,

Coller, H., Loh, M.L., Downing, J.R., Caligiuri, M.A., Bloomfield, C.D., and

Lander, E.S. (1999). Molecular classification of cancer: class discovery and

class prediction by gene expression monitoring. Science 286, 531-537.

Gorge, E., Brandt, S., and Werner, D. (1994). Uptake and metabolism of 2,4,6 TNT in

higher plants. Environ. Sci. Pollut. Res. 1, 229-233.

Gygi, S.P., Rochon, Y., Franza, B.R., and Aebersold, R. (1999). Correlation between

protein and mRNA abundance in yeast. Mol. Cell. Biol. 19, 1720-1730.

Haag, E., and Raman, V. (1994). Effects of primer choice and source of Taq DNA

polymerase on the banding patterns of differential display RT-PCR.

Biotechniques 2, 226-228.

Hannink, N., Rosser, S.J., French, C.E., Basran, A., Murray, J.A.H., Nicklin, S., and

Bruce, N.C. (2001). Phytodetoxification of TNT by transgenic plants expressing

a bacterial nitroreductase. Nature Biotechnology 19, 1168-1172.

Harvey, S.D., Fellows, R.J., Cataldo, D.A., and Bean, R.M. (1991). Fate of the

explosive hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) in soil and

bioaccumulation in bush bean hydroponic plants. Environ. Toxicol. Chem. 10,

845-855.

Haselwandter, M.M. (1983). Physiological Plant Ecology Vol. 3. Responses to the

Chemical and Biological Environment. (Berlin: Springer-Verlag).

Hawari, J., Halasz, A., Beaudet, S., Paquet, L., Groom, C., Ampleman, G., and

Thiboutot, S. (1999). Biotransformation of 2,4,6-trinitrotoluene with

Phanerochaete chrysosporium in agitated cultures at pH 4.5. Appl. and Environ.

Microbiol. 65, 2977-2986.

Hedge, R.S., and Fletcher, J.S. (1996). Influence of plant growth stage and season on

the release of root phenolics by mulberry as related to the development of

phytoremediation technology. Chemosphere 23, 2471-2479.

Hodgson, J., Rho, D., Guiot, S.R., Ampleman, G., Thiboutot, S., and Hawari, J.

(2000). Tween 80 enhanced TNT mineralization by Phaerochaete chrysosporium.

Can. J. Microbiol. 46, 110-118.

Hughes, J.B., Shanks, J.V., Vanderford, M., Lauritzen, J., and Bhadra, R. (1997).

Transformation of TNT by aquatic plants and plant tissue cultures. Environ. Sci.

Technol. 31, 266-271.

Jarvis, A.S., McFarland, V.A., and Honeycutt, M.E. (1998). Assessment of the

effectiveness of composting for the reduction of toxicity and mutagenicity of

explosive-contaminated soil. Ecotoxicol. and Environ. Safety 39, 131-135.

Jerger, D.E., and Woodhull, P.M. (2000). Applications and costs for biological

treatment of explosives-contaminated soils in the U.S. In Biodegradation of

nitroaromatic compounds and explosives, J.C. Spain, J.B. Hughes, and H.-J.

Knackmuss, eds (Boca Raton: CRC Press), pp. 395-424.

Kaplan, A.S., Berghout, C.F., and Peczenik, A. (1965). Human intoxication from

RDX. Arch. Environ. Health 10, 877-883.

Ketel, W.B., and Huyhes, J.R. (1972). Toxic encephalopathy with seizures secondary to

ingestion of composition C-4: A clinical and electroencephalographic study.

Neurology 22, 871-876.

Khan, J., Simon, R., Bittner, M., Chen, Y., Leighton, S.B., Pohida, T., Smith, P.D.,

Jiang, Y., Gooden, G.C., Trent, J.M., and Meltzer, P.S. (1998). Gene

expression profiling of alveolar rhabdomyosarcoma with cDNA microarrays.

Cancer Res. 58, 5009-5013.

Klein, W., and Scheunert, I. (1982). Bound pesticide residues in soil, plants, and food

with particular emphasis on the application for nuclear techniques. IEIA-SM-

263/38 Interatomic Energy Agency.

Larson, S.L. (1997). Fate of explosive contaminants in plants. Ann. N.Y. Acad. Sci. 829,

195-201.

Larson, S.L., Weiss, C.A., Escalon, B.L., and Parker, D. (1999a). Increased extraction

efficiency of acetonitrile/water mixtures for explosives determination in plant

tissues. Chemosphere 38, 2153-2162.

Larson, S.L., Jones, R.P., Escalon, B.L., and Parker, D. (1999b). Classification of

explosives transformation products in plant tissue. Environ. Toxicol. Chem. 18,

1270-1276.

Lashkari, D.A., DeRisi, J.L., McCusker, J.H., Namath, A.F., Gentile, C., Hwang,

S.Y., Brown, P.O., and Davis, R.W. (1997). Yeast microarrays for genome wide

parallel genetic and gene expression analysis. Proc. Natl. Acad. Sci. USA 94,

13057-13062.

Lauritzen, J. (1998). M.S. thesis. Rice University, Houston.

Levine, B.S., Furedi, E.M., Gordon, D.E., Barkley, J.J., and Lish, P.M. (1990a).

Toxic interactions of the munitions compounds TNT and RDX in F344 rats. Fund.

and Appl. Toxicol. 15, 373-380.

Lewis, R.J. (1992). Sax's Dangerous Properties of Industrial Materials. (New York: Von

Nostrand Reinhold).

Liang, P., and Pardee, A.B. (1992). Differential display of eukaryotic messenger RNA

by means of the polymerase chain reaction. Science 257, 967-971.

Lish, P.M., Levine, B.S., Furedi-Machacek, E.M., Sagartz, E.M., and Rac, V.S.

(1984). Determination of the Chronic Mammalian Toxicological Effects of RDX:

Twenty-four Month Chronic Toxicity/carcinogenicity Study of Hexahydro-1,3,5-

trinitro-1,3,5-triazine (RDX) in the B6C3FI Hybrid Mouse. Phase VI. Vol. I.

Contract No. DAMD 17-79-C-9161. U.S. Army Medical Research and

Development Command. DTIC AD-A160774.

Lynch, J.M. (1982). Interactions between bacteria and plants in the root environment. In

Bacteria and Plants. Soc. Appli. Bacteriol. Symp. Ser., M.E. Rhodes-Roberts and

F.A. Skinner, eds (London: Academic), pp. 1-23.

Matz, M.V., and Lukyanov, S.A. (1998). Different strategies of differential display:

areas of application. Nucl. Acids Res. 26, 5537-5543.

McCormick, N.G., Cornell, J.H., and Kaplan, A.M. (1981). Biodegradation of

hexahydro-1,3,5-trinitro-1,3,5-triazine. Appl. and Environ. Microbiol. 42, 817-

Meagher, R.B. (2000). Phytoremediation of toxic elemental and organic pollutants. Curr.

Opin. in Plant Biol. 3, 153-162.

Medina, V.F., Rivera, R., Larson, S., and McCutcheon, S.C. (1996).

Phytoremediation: modeling of TNT and its breakdown products. Soil

Groundwater Cleanup Feb/Mar, 19-24.

Meltzer, P.S. (2001). Large-scale genome analysis. In Bioinformatics: A Practical Guide

to the Analysis of Genes and Proteins. 2nd ed., A.D. Baxevanis, and Ouellette,

B.F.F., ed (New York City, NY: John Wiley and Sons, Inc.), pp. 393-412.

Michels, J., and Gottschalk, G. (1995). Pathway of 2,4,6-trinitrotoluene (TNT)

degradation by Phanerochaete chrysosporium. In Biodegradation of nitroaromatic

compounds, J.C. Spain, ed (New York, NY: Plenum Press), pp. 135-149.

Montpas, S., Samson, J., Langlois, E., Lei, J., Piche, Y., and Chevenert, R. (1997).

Degradation of 2,4,6-trinitrotoluene by Serratia marcensens. Biotechnol. Letters

19, 291-294.

Nuwaysir, E.F., Bittner, M., Trent, J., Barrett, J.C., and Afshari, C.A. (1999).

Microarrays and toxicology: the advent of toxicogenomics. Mol. Carcin. 24, 153-

Oh, K., and Kim, Y. (1998). Degradation of explosive 2,4,6-trinitrotoluene by s-triazine

degrading bacterium isolated from contaminated soil. Bull. Environ. Contam.

Toxicol. 61, 702-708.

Palazzo, A.J., and Leggett, D.C. (1986). Effect and disposition of TNT in terrestrial

plants. J. Environmental Quality 15, 49-52.

Palmer, W.G., Small, M.J., Dacre, J.C., and Eaton, J.C. (1996). Toxicology and

Environmental Hazards. P.L. Marinkas. (New York: Nova Science Publisher,

Inc.).

Preuss, A., Fimpel, J., and Dickert, G. (1993). Anaerobic transformation of 2,4,6-

trinitrotoluene (TNT). Arch. Microbiol 159, 345-353.

Rieble, S., Joshi, D.K., and Gold, M. (1994). Aromatic nitroreductase from the

basidiomycete Phanerochaete chrysosporium. Biochem. Biophys. Res. Commun.

205, 28-304.

Ross, D.T., Scherf, U., Eisen, M.B., Perou, C.M., Rees, C., Spellman, P., Iyer, V.,

Jeffrey, S.S., Van de Rijn, M., Waltham, M., Pergamenschikov, A., Lee, J.C.,

Lashkari, D., Shalon, D., Myers, T.G., Weinstein, J.N., Botstein, D., and

Brown, P.O. (2000). Systematic variation in gene expression patterns in human

cancer cell lines. Nat. Genet. 24, 227-235.

Rugh, C.L., Wilde, D., Stack, N.M., Thompson, D.M., Summer, A.O., and Meagher,

R.B. (1996). Mercuric ion reduction and resistance in transgenic Arabidopsis

thaliana plants expressing a modified bacterial merA gene. Proc. Natl. Acad. Sci.

USA 93, 3182-3187.

Ryon, M.G., Pal, B.C., Talmage, S.S., and Ross, R.H. (1984). Database Assessment of

the Health and Environmental Effects of Munition Production Waste Products.

Final Report, Army Project Order 83PP3802. ORNL-6018, Oak Ridge National

Laboratory, Oak Ridge, TN. DTIC AD-A145417.

Sandermann, J.H. (1994). Higher plant metabolism of xenobiotics: the 'green liver'

concept. Pharmacogenetics 4, 225-241.

Scheidemann, P.A., Klunk, A., Sens, C., and Werner, D.J. (1998). Species dependent

uptake and tolerance of nitroaromatic compounds by higher plants. Plant Physiol.

152, 242-247.

Scherf, U., Ross, D.T., Waltham, M., Smith, L.H., Lee, J.K., Tanabe, L., Kohn,

K.W., Reinhold, W.C., Myers, T.G., Andrews, D.T., Scudiero, D.A., Eisen,

M.B., Sausville, E.A., Pommier, Y., Botstein, D., Brown, P.O., and Weinstein,

J.N. (2000). A gene expression database for the molecular pharmacology of

cancer. Nat. Genet. 24, 236-244.

Schnell, S., and Schinck, B. (1991). Anaerobic aniline degradation via reductive

deamination of 4-aminobenzoyl-CoA in Desulfobacterium anilini. Arch.

Microbiol. 155, 183-190.

Schnoor, J.L. (1997). Phytoremediation. Technology Evaluation Report TE-97-01.

National Environmental Technology Applications Center, Pittsburgh, PA.

Schroder, P., and Collins, C. (2002). Conjugating enzymes involved in xenobiotic

metabolism or organic xenobiotics in plants. Int. Journ. of Phytoremed. 4, 247-

Schwab, A.P., and Banks, M.K. (1994). Biologically mediated dissipation of

polyaromatic hydrocarbons in the root zone. In Bioremediation through

Rhizosphere Technology, T.A. Anderson and J.R. Coats, eds (Washington, D.C.:

American Chemical Society.

Shimp, J.F., Tracy, J.C., Davis, L.C., Lee, E., and Huang, W. (1993). Beneficial

effects of plants in the remediation of soil and groundwater contaminated with

organic materials. Environ. Sci. Technol. 23, 41-77.

Smock, L.A., Stoneburger, D.L., and Clark, J.R. (1976). The toxic effects of

trinitrotoluene (TNT) and its primary degradation products on two species of

algae and the fathead minnow. Water Res. 10, 537-543.

Sompayrac, L., Jane, S., Burn, T.C., Tene, D.G., and Danna, K.J. (1995).

Overcoming limitations of the mRNA differential display technique. Nucl. Acids

Res. 23, 4738-4739.

Stahl, J.D., and Aust, S.D. (1995). Biodegradation of 2,4,6-trinitrotoluene by white rot

fungus Phanerochaete chrysosporium. In Biodegradation of nitroaromatic

compounds and explosives, J.C. Spain, ed (Boca Raton: Plenum Press).

Sun, Y., Hegamyer, G., and Colburn, N.H. (1994). Molecular cloning of five

messenger RNAs differentially expressed in preneoplastic or neoplastic JB6

mouse epidermal cells: one is homologous to human tissue inhibitor of

metalloproteinases-3. Cancer Res. 5, 1139-1144.

Tan, E.L., Hoe, C.H., Griest, W.H., and Tyndall, R.L. (1992). Mutagenicity of

trinitrotoluene and its metabolites formed during composting. Journ. of Toxicol.

and Environ. Health 36, 165-175.

The Arabidopsis Genome Initiative. (2000). Analysis of the genome sequence of the

flowering plant Arabidopsis thaliana. Nature 408, 796-815.

Thompson, P.L., Ramer, L., and Schnoor, J.L. (1998). Uptake and transformation of

TNT by hybrid poplar trees. Environ. Sci. Technol. 32, 975-980.

Thompson, P.L., Ramer, L., and Schnoor, J.L. (1999). Hexahydro-1,3,5-trinitro-1,3,5-

triazine translocation in poplar trees. Environ. Toxicol. Chem. 18, 279-284.

Traxler, R.W., Wood, E., and Delaney, J.M. (1974). Bacterial degradation of alpha-

TNT. Dev. Ind. Microbiol. 16, 71-76.

Vanderford, M. (1996). Aquatic phytotransformation of trinitrotoluene in contaminated

media (Houston: Rice University).

Vanderford, M., Shanks, J.V., and Hughes, J.B. (1997). Phytotransformation of

trinitrotoluene (TNT) and distribution of metabolic products in Myriophyllum

aquaticum. Biotech. Lettters 19, 277-280.

Velculescu, V.E., Zhang, L., Vogelstein, B., and Kinzler, K.W. (1995). Serial analysis

of gene expression. Science 270, 484-487.

Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Hornes, M., Frijters,

A., Pot, J., Peleman, J., Kuiper, M., and Zabeau, M. (1995). AFLP: a new

technique for DNA fingerprinting. Nucl. Acids Res. 23, 4407-4414.

Walton, B.T., Hoylman, A.M., Perez, M.M., Anderson, T.A., Johnson, T.R., Guthde,

T.E., and Christman, R.F. (1994). Rhizosphere microbial community as a plant

defense against toxic substances in soils. In Bioremediation Through Rhizosphere

Technology, T.A. Anderson, Coats, J.R., eds (Washington, DC: Am. Chem. Soc.),

pp. 82-92.

Watanabe, M.E. (2001). Can bioremediation bounce back? Nat. Biotechnol. 19, 1111-

Wayment, D.G., Bhadra, R., Lauritzen, J., Hughes, J., and Shanks, J.V. (1999). A

transient study of formation of conjugates during TNT metabolism by plant

tissues. Int. Journ. of Phytoremed. 1, 227-239.

Wodicka, L., Dong, H., Mittman, M., Ho, M.H., and Lockhart, D.J. (1997). Genome-

wide expression monitoring in Saccharomyces cerevisiae. Nat. Biotechnol. 15,

1359-1367.

Wolfe, N.L., Ou, T.Y., Carriera, L., and Gunnison, D. (1994). Alternative methods for

biological destruction of TNT: a preliminary feasibility assessment of enzymatic

degradation. Technical Report IRRP-94-3. U.S. Army Corps of Engineers

Waterways Experiment Station, Vicksburg, MS.

Won, W.D., Disalvo, L.H., and Ng, J. (1976). Toxicity and mutagenecity of 2,4,6-

trinitrotoluene and its microbial metabolites. Appl. and Environ. Microbiol. 31,

576-580.

Yinon, J. (1990). Toxicity and metabolism of explosives. (Ann Arbor, MI: CRC Press).

Zhang, J.S., Duncan, E.L., Chang, A.C.M., and Reddel, R.R. (1998). Differential

display of mRNA. Mol. Biotechnol. 10, 155-165.

CHAPTER 2

SAGE ANALYSIS OF TRANSCRIPTOME RESPONSES IN ARABIDOPSIS

THALIANA ROOTS EXPOSED TO TNT1

______________________________

1Ekman, D.R., Lorenz, W.W., Przybyla, A.E., Wolfe, N.L., and Dean, J.F.D. Submitted

to Plant Physiology.

ABSTRACT

Serial Analysis of Gene Expression (SAGE) was used to profile transcript levels

in Arabidopsis thaliana roots and assess their responses to 2,4,6-trinitrotoluene (TNT)

exposure. SAGE libraries representing control and TNT-exposed seedling root

transcripts were constructed, and each was sequenced to a depth of roughly 30,000 tags.

More than 19,000 unique tags were identified overall. The second most highly induced

tag (27-fold increase) represented a glutathione S-transferase. Several cytochrome P450

enzymes, as well as an ABC transporter, were highly induced by TNT exposure. These

analyses also revealed an oxidative stress response upon TNT exposure. Although some

increases were anticipated in light of current models for xenobiotic metabolism in plants,

evidence for unsuspected conjugation pathways was also noted. Identifying

transcriptome-level responses to TNT exposure will better define the metabolic pathways

plants use to detoxify this xenobiotic compound, which should help improve

phytoremediation strategies directed at TNT and other nitroaromatic compounds.

(Note: the datasets from these analyses are available at the ncbi Gene Expression

Omnibus (GEO) at: http://www.ncbi.nlm.nih.gov/geo/)

INTRODUCTION

Soil and groundwater at sites throughout the U.S. and Europe were contaminated

in the past century by manufacturing, processing, and storage of explosives, such as

2,4,6-trinitrotoluene (TNT)1. Unlike many other nitroaromatic compounds, including

pesticides and various feedstock chemicals, the energetic nitroaromatics (Figure 2.1) are

highly resistant to degradation and may persist in the environment for decades. Certain

plant species have the ability to accumulate TNT from their surroundings and, thus, offer

a potential means for removing these compounds from the environment2. Unfortunately,

few of these species are capable of tolerating the contamination levels common to sites

most in need of remediation. Transgenic plants have been developed with enhanced

abilities to tolerate and remove TNT from soil under laboratory conditions3,4. However,

lack of information on the biochemical mechanisms involved in TNT uptake and

metabolism limits our ability to modify and adapt plants specifically for this task.

To obtain a more complete picture of the metabolic processes plants employ to

cope with nitroaromatic agents, Serial Analysis of Gene Expression (SAGE)5,6 was used

to identify transcriptome-level responses in A. thaliana seedling roots exposed to TNT.

Although this technique has been widely used to study gene expression changes in human

cancers7, only two published reports describe the use of SAGE in plants: a study of gene

expression in rice seedlings8, and an examination of wood formation in loblolly pine9.

The identification of plant genes and pathways responding to TNT exposure at the

transcriptional level should facilitate a more reasoned approach to the development of

plants better suited for use in phytoremediation of TNT-contaminated sites.

RESULTS

SAGE Libraries.

SAGE libraries representing transcripts expressed in Arabidopsis root tissues

grown in the absence or presence (15 mg/L) of TNT were sequenced to characterize

about 30,000 tags from each. The datasets from these analyses are available online in

spreadsheet format (http://www.arches.uga.edu/~jeffdean/SAGE/SAGEData.html). Of

the 32,203 tags characterized from the library for TNT-treated tissue, 12,005 represented

unique transcripts, but 7,900 of these were singletons. Similarly, 12,719 unique tags

were encountered amongst the 31,973 characterized in the control library, with 8,322 of

these representing singletons. A double-reciprocal plot of the total unique tags identified

versus the total number of tags sequenced after each sequencing run was used to estimate

the rate of transcript discovery (Fig. 2.2). From this, transcriptome sizes of

approximately 21,000 and 15,000 were estimated for control and TNT-treated root

tissues, respectively. Only 25% of all unique tags (5,084 out of 19,640) were detected in

both libraries.

BLAST Analyses.

Of the 1,045 most abundant tags in both SAGE libraries (which included all tags

seen 10 or more times amongst the 64,176 tags characterized) about 70% (739), could be

matched to a single model gene in the AGI (Arabidopsis Genome Initiative) database.

Most of the remaining tags (233) matched sequences found one or more times among all

Arabidopsis sequences deposited in GenBank. Many such tags were found in expressed

sequence tags (ESTs), but were not positioned adjacent to the CATG sequence closest to

the 3’ end of the transcript. This suggests they might have been generated from

alternatively spliced transcripts. A good example of this was the fourth most abundant tag

in the control tissue library (AGGTCTTGGT, counted 134 times). This sequence appears

only one time in the Arabidopsis genome, and falls immediately 3’ of the penultimate

CATG site on the annotated transcript of gene At3g09260 (β-glucosidase). The full-

length transcript for this gene annotated in the AGI database was responsible for second

most abundant tag in the control library (ATTTGCCAGA, counted 286 times) and the

fourth most abundant tag in the TNT-treatment library (counted 259 times).

Of the remaining unidentified tags in the top 1,045, one that did not correspond to

anything listed in the AGI database (AGTAACGATA) matched a sequence found on the

Arabidopsis mitochondrial genome, and numerous ESTs incorporating this sequence

have been deposited in GenBank. About 5% of the tags (57) matched more than one

model gene, while a few tags (14) contained at their 3’ end a contiguous stretch of “A”

residues suggesting that they incorporated part of a poly(A) tail. Two tags, one of which

(TCCCCGTACA) was the 37th most abundant tag overall, could not be matched to any

Arabidopsis sequences in GenBank.

Differentially Expressed Transcripts.

TNT treatment induced an apparent increase of at least five-fold in 242 tags,

while 287 tags decreased in abundance at least five-fold in response to this treatment. For

tags that were relatively abundant in one library but not observed in the other, a minimum

change in expression level was estimated by assuming that a single copy of the tag was

found in the library from which it was absent. Many of the tags displaying the greatest

induction in response to TNT exposure represented gene products known to be involved

in plant responses to oxidative stress, such as monodehydroascorbate reductase and

phospholipid-hydroperoxide glutathione peroxidase (Table 2.1). The identity of the tag

most highly induced by TNT remains uncertain. Its sequence appears twice on

Chromosome 5, and in one of these instances the tag lies in the middle of a

computationally predicted transcript that has never been isolated. However, a third

occurrence of the sequence resides on Chromosome 2 at the penultimate CAGT site on

the transcript for LKP2 (At2g18910), a signaling protein involved in the Arabidopsis

circadian clock10. The evidence suggests the most likely source of this tag to be a splice

variant of the LKP2 transcript. In contrast, the second most highly induced tag, which

increased nearly 28-fold in TNT-treated roots, definitely represented a glutathione S-

transferase (GST). Several of the induced tags represented cytochrome P450s, a large

family of enzymes known to be involved in the detoxification of xenobiotic compounds

by plants and other organisms (Table 2.2). As further indication of the stress imposed

upon the seedlings by TNT exposure, tags for various transporter proteins, transcription

factors, signal cascade proteins, and heat shock proteins were also among those most

highly induced by TNT exposure. Other strongly induced tags representing gene products

known to be involved in detoxification reactions or in the protection of plant cells against

oxidative stress are noted in Table 2.3. Prominent among the enzymes likely to protect

against oxidative stress are GSH-dependent dehydroascorbate reductase, 1,4-

benzoquinone reductase, peptide methionine sulfoxide reductase, and glutathione

reductase.

Tags that showed the greatest decrease upon exposure to TNT are listed in Table

2.4. The largest effect was seen for a tag representing AIR1A, a transcript that has been

associated with lateral root formation in response to auxin11. AIR1A and AIR1B, a

nearly identical transcript that appears to arise via alternative splicing, encode proteins

that are similar to a group of cell wall-plasma membrane connector proteins widely

distributed in plants12. However, neither of the AIR1 proteins harbors the N-terminal

proline- or glycine-rich extracellular domain thought to mediate cell wall interaction in

other members of the connector protein family. Interestingly, the AIR1B transcript is

represented by a different SAGE tag, which happens to be the tag showing the fourth

most precipitous decline in response to TNT. Another member of the cell wall-cell

membrane connector protein family is pEARLI-1, an Arabidopsis homolog that was also

severely repressed in response to the TNT treatment (Table 2.4).

TNT exposure also depressed levels of tags representing three different latex

protein homologs, another group of widely distributed proteins whose functions are as yet

unknown13. Other depressed tags included those for various transcription factors,

membrane channels, cytoskeletal elements, and ribosomal proteins.

Quantitative PCR.

SAGE has previously been shown to provide an accurate reflection of gene

expression levels for medium- and high-abundance transcripts14, but the induction or

repression of selected Arabidopsis transcripts in response to TNT exposure was

independently verified in this study using real-time quantitative PCR. Genes that SAGE

analysis suggested were induced (At3g28740; cytochrome P450), repressed (At2g36830;

aquaporin) or remained unaffected (At2g39460; 60S ribosomal protein L23A) by TNT

exposure were tested. As shown in Table 2.5, the quantitative PCR data showed general

agreement with the SAGE data.

DISCUSSION

A major goal of this study was to identify plant-specific enzymes and metabolic

pathways that might previously have been overlooked for their importance in conferring

tolerance to TNT [seedlings were grown in sterile liquid culture where microbial

influences could be avoided]. The tolerance of Arabidopsis to TNT under these

conditions (15 mg/L) was comparable to those seen in other plants grown under sterile

culture conditions4. For example, TNT was shown to induce chlorosis and cell death in

cultures of Anabena at a concentration of 10 mg/L15, while cell suspension cultures of

Datura innoxia were capable of tolerating TNT at up to 29 mg/L without an affect on

growth16. Although the liquid culture system used in this study exposed all portions of

the Arabidopsis seedlings to TNT, only the root tissues were sampled for SAGE analyses

because under conditions of normal terrestrial growth TNT and its derivatives tend to

remain associated with root tissues17. Thus, efforts to improve plant tolerance to TNT

would seem most likely to benefit from detailed study of the metabolic responses in these

tissues.

SAGE Analysis of the Arabidopsis Transcriptome.

The Arabidopsis genome has been estimated to harbor approximately 25,500

genes18, but given the possibilities for alternative splicing and polyadenylation during

transcript maturation in eukaryotes, the transcriptional space of Arabidopsis should be

substantially larger. This inference was supported by SAGE results identifying more

than 19,000 unique tags in a sampling of only 60,000 tags from root tissues alone. Just as

was seen in a similarly sized SAGE study of rat fibroblast cells19, the discovery of new

tags in the Arabidopsis SAGE libraries had not yet begun to level off at this depth of

sampling, and estimates suggested an eventual transcriptome size in the range of 35,000-

40,000 for the root tissues. However, as noted by Velculescu et al.20, to obtain a

reasonably complete sampling of a somewhat larger transcriptome (ca. 56,000

transcripts) using SAGE, one would need to sequence on the order of ca. 650,000 tags.

Thus, our data likely represent a low estimate of transcriptional capacity in Arabidopsis.

The size of this SAGE study was similar to a hypothetical situation (15,720 genes,

62,178 sampled tags) modeled by Stollberg et al.21 to estimate the maximum likelihood

of generating accurate transcript number and transcript copy frequency given different

levels of randomness and non-uniqueness in the transcriptome. Modeling suggested that

in a study of this size 1.5 - 6% of SAGE tags should represent two or more genes

depending on the non-randomness of sequences represented in the transcriptome. This

value matched closely with the 5% figure noted for the 1,045 most abundant tags in this

study. However, the modeling study did not take into account the further complexity that

might be introduced by variations in transcript maturation processes, such as the

alternative splicing. Adding further uncertainty to estimates of the Arabidopsis root

transcriptome size was the observation that among tags categorized as having ambiguous

origin, several matched the antisense sequences for SAGE tags predicted for certain

transcripts having numerous ESTs on deposit in GenBank. SAGE previously

demonstrated that antisense transcripts for certain genes were abundant in Caenorhabditis

elegans22 and Plasmodium falciparum23, and although not verified in this study, SAGE

tags from antisense transcripts have been verified by RT-PCR in stressed Arabidopsis

plants (G. May, personal communication). Perhaps this phenomenon is a reflection of the

up-regulation of endogenous reverse transcriptases, such as that encoded by At2g166680,

which was induced seven-fold by TNT exposure (Table 2.1).

Multiphase Mechanisms of TNT Detoxification in Arabidopsis.

Studies of the mechanisms plants use to metabolize herbicides and other

xenobiotics have previously pointed to a multiphase process for detoxification24,25,26.

Following entrance of the compound into the cell, transformation reactions introduce

chemical substituents amenable to conjugation, for example, hydroxyl groups added via

enzyme-catalyzed oxidation reactions involving cytochrome P450 enzymes and other

mixed function oxidases. The modified compound is subsequently conjugated to

glutathione (GSH) or any of several six-carbon sugars through the action of glutathione

S-transferases (GSTs) or UDP:glucosyltransferases, respectively. In the third phase of

the process, conjugates are sequestered in the vacuole, where they may be inactivated by

further modifications or enter into degradation pathways, or secreted into the apoplasm

where they may be covalently coupled into the cell wall.

SAGE analysis indicates that this multiphase process also functions in the

metabolism of TNT by Arabidopsis. Although this has been suggested previously for

two other plants, Catharanthus roseus and Myriophyllum aquaticum, based on mass-

balance studies of TNT disappearance and metabolite production in axenic cultures27,28,

the enzymes most likely involved in the process have not previously been determined.

The apparent induction of several cytochrome P450 genes by TNT exposure supports

oxidation as the initial transformation step in TNT metabolism. Oxidative transformation

of TNT in plants has been suggested from analyses of metabolic products28, but reductive

metabolites, such as 2-amino-4,6-dinitrotoluene or 4-amino-2,6-dinitrotoluene, have been

recovered most often in mass-balance studies of plant cultures exposed to 14C-TNT.

Typically, however, the fraction of TNT derivatives recoverable from such studies as

small molecules has been only a small percentage of the added TNT, and most of the

isotopic label recovered from such studies remained in unidentified conjugates or bound

to high molecular weight materials. This suggests that after oxidation of TNT,

subsequent conjugation reactions occur too quickly to allow for routine measure of the

oxidized intermediates.

While hydroxylation is an effective means of modifying xenobiotic compounds

for subsequent conjugation, it rarely serves to decrease toxicity of the parent compound

and can sometimes make the compound more toxic. In such cases, hydroxylating

enzymes must be coordinately expressed with conjugating enzymes that will decrease the

toxic nature of the derivatives as quickly as possible. The SAGE data suggest that

glutathione S-transferases were the enzymes shouldering primary responsibility for

conjugation reactions involving TNT metabolites. Although previous studies have not

provided evidence for the conjugation of glutathione to TNT, its conjugation to

herbicides and other phytotoxins is well known. In fact, detoxification of herbicides by

glutathione conjugation is used advantageously in agriculture where compounds that

elevate glutathione and glutathione S-transferase levels, known as “safeners,” are applied

to crops prior to herbicide application29. Many safeners are aromatic compounds with

nitrogen-containing functional groups, not unlike TNT, and their metabolism in plants

has been relatively well studied. Enhanced glucosylation reactions in response to safener

treatment have also been noted30, and may explain the TNT induction of a SAGE tag

representing sucrose-UDP glucosyltransferase. From these observations it would seem

that many of the mechanisms known to function in safener metabolism may be directly

applicable in understanding the response of Arabidopsis to TNT exposure.

On the other hand, the seven-fold induction of anthranilate N-benzoyltransferase

(Table 2.3), an enzyme that catalyzes the first committed step of a phytoalexin

biosynthetic pathway and that has been shown to be adept at utilizing hydroxycinnamoyl-

CoA esters to modify anthranilate31, may suggest a conjugation pathway for TNT

derivatives that has not been previously suspected. Bhadra et al.28 found that oxidation of

the methyl group converted nearly 20% of the TNT added to plant cultures to various

benzoyl derivatives. As benzoate:CoA ligase is a member of an enzyme family that

includes 4-coumarate:CoA ligase32, the near three-fold induction of 4-coumarate:CoA

ligase seen in the TNT-treated roots may reflect a mechanism for activating oxidized

TNT derivatives prior to anthranilate conjugation.

Conjugation of glutathione with cellular toxins is generally thought to confer

increased solubility, decreased toxicity, and increased transport competency on the toxin

derivatives33. The transport of GSH conjugates across membranes is an ATP-dependent

process handled by ABC transporter proteins34,35,36. As noted in Table 2.3, SAGE

detected five-fold increase in an ABC transporter (At3g53480) upon TNT exposure. This

ABC transporter is of a class related to the S. cerevisiae PDR5 gene and, as such, has

been suggested to have potential function as a toxin-conjugate efflux pump37. Treatment

of liquid-grown Arabidopsis seedlings with the nitroaromatic cytotoxin, 1-chloro-2,4-

dinitrobenzene (CDNB), elicited similar increases in transcript levels for three other ABC

transporters, including the AtMRP1 (At1g30400) and AtMRP4 (At2g47800), that are

thought to handle transport of glutathione conjugates and anthocyanins into the vacuole38.

In addition to undergoing oxidation reactions, the TNT taken up by plants is also

subject to reduction reactions that target the aromatic nitro-groups, as evidenced by the

amino, dinitro-derivatives of TNT identified in plant tissues by Bhadra et al.28. SAGE

suggests that such reduction reactions may be catalyzed by NADPH-dependent

flavoenzymes, such as 12-oxophytodienoate reductase (At1g76680), which was induced

10-fold by TNT treatment. Similar in sequence to the yeast Old Yellow Enzyme (OYE),

12-oxophytodienoate reductase is related to the nitrate ester reductases39, and another

member of this family, pentaerythritol trinitrate (PETN) reductase, has been shown to

increase TNT tolerance when expressed in transgenic tobacco plants3. Thus, 12-

oxophytodienoate reductase may prove a useful target for over-expression experiments

aimed at increasing plant tolerance to TNT and other nitroaromatic compounds.

Although there is no published evidence for mineralization of TNT by plants, the

strong induction by TNT of carbamoyl phosphate synthetase (23-fold) suggests that

nitrogen metabolism changed in the treated plants, possibly reflecting an unappreciated

mechanism for removing nitrogen from the aromatic ring. Isotopic labeling studies have

been so far limited to carbon-14, but tracking of labeled nitrogen might help to further

clarify the metabolic pathways in plants for TNT.

Gene Repression in Response to TNT.

Inference of metabolic function for genes whose expression is induced by stress

can be relatively straightforward given sufficient information on the activity of the

encoded proteins. However, understanding the down-regulation of other genes under the

same conditions of stress is much more difficult since their repression may be the indirect

outcome of regulatory shifts necessary to induce protective genes. Among the transcripts

most repressed by TNT exposure were several whose products have previously been

associated with the normal structure and function of plant cell walls, e.g., the cell wall-

plasma membrane linker proteins, AIR1a, AIR1b and pEARLI-1, an extensin, and an

arabinogalactan protein (Table 2.4). As a group, latex protein homologs were also highly

repressed. Although these proteins have no known function, they are generally thought

to accumulate in vacuoles. Thus, it is possible that their repression reflects changes in

vacuole metabolism brought about by sequestration of TNT-conjugates in this organelle.

Changes in vacuolar metabolism might also explain the repression of aquaporin and

membrane channel proteins noted in Table 2.4 if these particular family members are

associated with the tonoplast.

This study was undertaken to improve our understanding of the mechanisms

involved in plant tolerance and metabolism of xenobiotic compounds, particularly TNT.

Results from these experiments suggest the involvement of previously unappreciated

enzymes and, at the same time, strengthen some existing theories of how plants cope with

toxic compounds. Identification of the genes involved in the metabolism of TNT should

provide for focused engineering attempts to create plants better suited to remediation.

EXPERIMENTAL PROTOCOL

Plant Material, Growth Conditions, and Root Tissue Isolation.

A. thaliana ecotype Columbia seeds (WT-2, Lehle Seeds, Round Rock, TX) were

surface-sterilized and placed in sterile Murashige and Skoog liquid medium prepared

according to the manufacturer’s protocol (Invitrogen Life Technologies, Carlsbad, CA).

Plants were grown for 14 days at 250C under a 16-hour photoperiod with constant

shaking at 85 rpm in a growth chamber. TNT was obtained from the U.S. Army Center

for Environmental Health Research (Fort Detrick, MD). Toxicity was assessed by adding

TNT from a stock solution in DMSO to yield final concentrations of 0, 5, 10, 15, 20, 25,

30, and 40 mg/L in MS medium, and cultures were run in triplicate. The cultures were

returned to the growth chamber for five days, during which time the seedlings in each

flask were examined for signs of stress (leaf chlorosis and necrosis). A final

concentration of 15 mg/L TNT was judged to produce notable stress in the plants without

causing death. Root tissues for SAGE library construction were isolated from seedlings

grown 14 days in liquid MS medium and dosed with TNT to a final concentration of 15

mg/L. Control tissues were isolated from seedlings grown under the same conditions and

dosed with an equivalent volume of DMSO. The seedlings were grown in the presence

of TNT or DMSO for 24 hours, after which they were submerged briefly in dH2O to

remove excess medium, and excised roots were immediately frozen in liquid nitrogen.

To obtain sufficient biomass, five flasks of seedlings (70-80 seeds in 200 ml medium in a

500 ml flask) received each treatment, and treatments were replicated on three separate

occasions. Root tissues were pooled by treatment and stored at –800C prior to RNA

extraction.

RNA Isolation and cDNA Synthesis.

Total RNA from root tissues was extracted using the LiCl precipitation technique

of Chang et al.40. Poly(A) RNA was isolated from total RNA using Dynabeads oligo-

dT(25) magnetic beads (Dynal Biotech, Lake Success, NY) at a ratio of 0.25 mg of total

RNA per 250 µL of Dynabeads and following the manufacture’s instructions. Double-

stranded cDNA was synthesized from 5 mg of poly(A) RNA using the Superscript

Choice cDNA synthesis kit (Invitrogen Life Technologies) and following the

manufacturer’s protocol, except for the substitution of a 5’-biotin dT(18) primer in the

first-strand reaction.

SAGE Library Construction.

SAGE libraries were constructed according to the SAGE Detailed Protocol,

Version 1.0c6, a brief description of which follows. Biotinylated cDNAs from each tissue

sample were bound to streptavidin-coated magnetic beads and digested with NlaIII, a

restriction enzyme recognizing the four-base sequence, CATG (anchoring enzyme).

DNA released by this digestion was washed away and the beads, with the adherent 3’

ends of each cDNA, were split into two pools. Linkers containing a binding site for

BsmF1 (a Type-II restriction endonuclease – the tagging enzyme), but different sites for

PCR primers, were ligated to the NlaIII cleavage site at the 5’ ends of the bead-bound

cDNA fragments in each pool. Both pools of cDNAs were digested with BsmF1 to

release SAGE tags from the beads, after which the pools were combined, and 102 bp

linker-flanked ditags were formed by blunt-end ligation. Following amplification of the

ditags by PCR, the linkers were removed by NlaIII digestion, and ditags were ligated to

form concatemers. The concatemers were subsequently size-fractionated, ligated into the

pZero vector (Invitrogen Life Technologies), cloned, and sequenced.

SAGE Data Analysis.

Sequence files were compiled and analyzed using the SAGE Software, ver. 3.03,

provided by Dr. Kenneth Kinzler (Johns Hopkins University, Baltimore, MD). Tags

containing linker sequences and repeated ditags were excluded prior to analysis. Because

the library representing TNT-treated roots was sequenced to a slightly greater extent

(32,203 tags for TNT treatment vs. 31,973 tags for the control), values for the control

library tags were normalized prior to making comparisons of relative gene expression.

Ratios were used to compare the relative expression of tags between the two libraries

(e.g., TNT/control), and in instances where a particular tag was absent from a library, a

value of 1 was substituted to avoid division by zero. Using the SAGE software, Monte

Carlo simulations were performed to estimate the statistical significance of any

differential expression. The null hypothesis for these analyses was that the abundance,

type, and distribution of transcripts were the same in both libraries.

Gene Identification.

To identify the genes from which tags were derived, each 10-base tag plus the 4-

base NlaIII recognition sequence was first compared against the Arabidopsis Gene

Initiative (AGI) database of model genes using the Patmatch analysis tool available on

The Arabidopsis Information Resource (TAIR) server (http://www.arabidopsis.org). If

the tag was found to match exactly the NlaIII site closest to the 3’ end of a model gene,

this identity was accepted for the tag. Tags that could not be found in the model gene

database were compared against all Arabidopsis sequences in GenBank using the same

Patmatch tool. Exact matches were annotated accordingly.

Quantitative PCR.

Total RNA from Arabidopsis plants grown under conditions identical to those

used to generate RNA for the SAGE studies was used to independently verify the

expression of selected genes by quantitative RT-PCR. Messenger RNA isolated using

Dynal Oligo dT25 magnetic beads served as template for single-stranded cDNA synthesis

using Taqman Reverse Transcription Reagents (Applied Biosystems, Foster City, CA).

Real-time fluorescent detection of RT-PCR products was performed using an ABI Prism

7700 sequence detection system and Sybr Green PCR Master Mix (Applied Biosystems).

The following PCR primers were designed for this study using Primer Express v. 1.0

(Applied Biosystems): cytochrome P450 (At3g28740) forward, 5’-

TTGATGCCTTTTGGGATTGG-3’ and reverse, 5’-CAAGGTCACTAGCCGTTGAGC-

3’; 60S ribosomal protein L23A (At2g39460) forward, 5’-

TCCAGACCAAGAAAGTGAACACA-3’ and reverse, 5’-

CATAGTCTGGTGTAAGCCTCACGT-3’; and aquaporin (At2g36830) forward, 5’-

GCTTCTCGGCTCCGTCG-3’ and reverse, 5’-GGCACAGCCAAGCCACC-3’.

Amplification reactions were carried out according to the manufacturer’s specifications

as follows: two minutes at 500C followed by a 10 min activation of the enzyme at 950C,

and 40 subsequent cycles consisting of 950C for 15 sec followed by 600C for 1 min. All

amplification reactions were run using a dilution series of cDNA from either control or

TNT-treated tissues using the same PCR master mix. Amplimers were checked for purity

and size by gel electrophoresis to ensure that the correct sequence was amplified.

Controls reactions omitting reverse transcriptase were run for all samples to ensure that

genomic contamination did not contribute to the amplified products.

Acknowledgements

This work was supported by NNEMS Fellowship U-91587201-0 from the U.S.

Environmental Protection Agency to D.R.E. We thank Arthur Karnaugh for help with

DNA sequencing protocols, and Caroline Stevens for assistance with data analysis.

Thanks also to MacArthur Long and Steve McCutcheon for help in initiating the project

and its continued support, and Jeff Dangl for comments on the manuscript.

REFERENCES

1. Walsh, M.E., Jenkins, T.F., Schnitcker, P.S., Elwell, J.W., & Stutz, M.H. USA

Cold Regions Research and Engineering Laboratory CRREL Special Report 93-5,

Hanover NH, pp 1-17 (1993).

2. Hannink, N.K., Rosser, S.J., & Bruce, N.C. Phytoremediation of explosives. Crit.

Rev. Plant Sci. 21, 511-538 (2002).

3. French, C. E., Rosser S.J., Davies, G.J., Nicklin, S., & Bruce, N.C.

Biodegradation of explosives by transgenic plants expressing pentaerythritol

tetranitrate reductase. Nat. Biotech. 17, 491-494 (1999).

4. Hannink, N., et al. Phytodetoxification of TNT by transgenic plants expressing a

bacterial nitroreductase. Nat. Biotech. 19, 1168-1172 (2001).

5. Velculescu, V.E., Zhang, L., Vogelstein, B., & Kinzler, K.W. Serial analysis

of gene expression. Science 270, 484-487 (1995).

6. Velculescu, V.E., et al. Characterization of the yeast transcriptome. Cell 88, 243-

251 (1997).

7. Riggins, G.J., & Strausberg, R.L. Genome and genetic resources from the Cancer

Genome Anatomy Project. Hum. Mol. Genet. 10, 663-667 (2001).

8. Matsumura, H., Nirasawa, S., & Terauchi, R. Technical advance: transcript

profiling in rice (Oryza sativa L.) seedlings using serial analysis of

gene expression. Plant J. 20, 719-726 (1999).

9. Lorenz, W.W., & Dean, J.F.D. SAGE Profiling and demonstration of differential

gene expression along the axial developmental gradient of lignifying xylem in

loblolly pine (Pinus taeda). Tree Physiol. 22, 301-10 (2002).

10. Schultz, T.F., Kiyosue, T., Yanovsky, M., Wada, M. & Kay, S.A. A role for

LKP2 in the circadian clock of Arabidopsis. Plant Cell 13, 2659-2670 (2001).

11. Neuteboom, L.W., et al. Isolation and characterization of cDNA clones

corresponding with mRNAs that accumulate during auxin-induced lateral root

formation. Plant Mol. Biol. 39, 273-287 (1999).

12. Goodwin, W., Pallas, J.A., & Jenkins, G.I. Transcripts of a gene encoding a

putative cell wall plasma membrane linker protein are specifically cold-induced in

Brassica napus. Plant Mol. Biol. 31, 771-781 (1996).

13. Stromvik, M.V., Sundararaman, V.P., & Vodkin, L.O. A novel promoter from

soybean that is active in a complex developmental pattern with and without its

proximal 650 base pairs. Plant Mol. Biol. 41, 217-231 (1999).

14. Evans, S.J., et al. Evaluation of Affymetrix gene chip sensitivity in rat

hippocampal tissue using SAGE analysis. Serial Analysis of Gene Expression.

Eur. J. Neurosci. 16, 409-13 (2002).

15. Pavlostathis, S.G., & Jackson, G.H. (1999). Biotransformation of 2,4,6-

trinitrotoluene in Anabaena sp. cultures. Environ. Toxicol. Chem. 18, 412-419

(1999).

16. Lucero, M.E., Mueller, W., Hubstenberger, J., Phillips, G.C., & O'Connell, M.A.

Tolerance to nitrogenous explosives and metabolism of TNT by cell suspensions

of Datura innoxia. In Vitro Cell Dev. Plant 35, 480-486 (1999).

17. Cataldo, D.A., Harvey, S.D., Fellows, R.J., Bean, R.M., & McVeety, B.D. An

Evaluation of the Environmental Fate and Behavior of Munitions Material (TNT,

RDX) in Soil and Plant Systems (Report AD-A223, 546). Pacific Northwest

Laboratories, Richland, WA (1989).

18. The Arabidopsis Initiative. Analysis of the genome sequence of the flowering

plant Arabidopsis thaliana. Nature 408, 796-815 (2000).

19. Madden, S., Galella, E., Zhu, J., Bertelsen, A., & Beaudry, G. SAGE transcript

profiles for p53-dependent growth regulation. Oncogene 15, 1079-1085 (1997).

20. Velculescu, V.E., et al. Analysis of human transcriptomes. Nat. Genet. 23, 387-

388 (1999).

21. Stöllberg, J., Urschitz, J., Urban, Z., & Boyd, C.D. A quantitative evaluation of

SAGE. Genome Res. 10,1241-1248 (2000).

22. Jones, S.J., et al. Changes in gene expression associated with developmental arrest

and longevity in Caenorhabditis elegans. Genome Res. 11, 1346-1352 (2001).

23. Patankar, S., Munasinghe, A., Shoaibi, A., Cummings, L.M., & Wirth, D.F. Serial

analysis of gene expression in Plasmodium falciparum reveals the global

expression profile of erythrocytic stages and the presence of anti-sense transcripts

in the malarial parasite. Mol. Biol. Cell. 12, 3114-3125 (2001).

24. Ishikawa, T. The ATP-dependent glutathione S-conjugate export pump. Tr.

Biochem. Sci. 17, 433-438 (1992).

25. Ishikawa, T., Li, Z-S., Lu, Y-P., & Rea, P.A. The GS-X pump in plant, yeast, and

animal cells: Structure, function, and gene expression. Biosci. Rep. 17, 189-207

(1997).

26. Kreuz, K., Tommasini, R., & Martinoia, E. Old enzymes for a new job -

Herbicide detoxification in plants. Plant Physiol. 111, 349-353 (1996).

27. Bhadra, R., Wayment, D.G., Hughes, J.B., & Shanks, J.V. Confirmation of

conjugation processes during TNT metabolism by axenic plant roots. Environ.

Sci. Technol. 33, 446-452 (1999a).

28. Bhadra, R., Wayment, D.G., Hughes, J.B., & Shanks, J.V. Characterization of

oxidation products of TNT metabolism in aquatic phytoremediation systems of

Myriophyllum aquaticum. Environ. Sci. Technol. 33, 3354-3361 (1999b).

29. Davies, J., & Caseley, J.C. Herbicide safeners: a review. Pestic. Sci. 55, 1043-

1058 (1999).

30. Kreuz, K. , Gaudin, J., Stingelin, J. & Ebert, E. Metabolism of

aryloxyphenoxypropanoate herbicide, CGA 184927, in wheat, barley and

maize:differential effects of the safener CGA 185072. Z Naturforschung 46, 901-

905 (1991).

31. Yang, Q., Reinhard, K., Schiltz, E., & Matern, U. Characterization and

heterologous expression of hydroxycinnamoyl/benzoyl-CoA : anthranilate N-

hydroxycinnamoyl/benzoyltransferase from elicited cell cultures of carnation,

Dianthus caryophyllus L. Plant Mol. Biol. 35, 777-789 (1997).

32. Beuerle, T., & Pichersky, E. Purification and characterization of

benzoate:coenzyme A ligase from Clarkia breweri. Arch. Biochem. Biophys. 400,

258-264 (2002).

33. Schröder, P., & Collins, C. Conjugating enzymes involved in xenobiotic

metabolism of organic xenobiotics in plants. Intl. J. Phytoremed. 4, 247-265

(2002).

34. Martinoia, E., Grill, E., Tommasini, R., Kreuz, K., & Amrhein, M. ATP-

dependent glutathione S-conjugate export pump in the vacuolar membrane of

plants. Nature 364, 247-249 (1993).

35. Li, Z-S., Zhao, Y., & Rea, P.A. Magnesium adenosine 5'-triphosphate-energized

transport of glutathione-S-conjugates by plant vacuolar membrane vesicles. Plant

Physiol. 107, 1257-1268 (1995).

36. Theodoulou, F.L. Plant ABC transporters. Biochim. Biophys. Acta 1465, 79-103

(2000).

37. Mitterbauer, R., & Adam, G. Saccharomyces cerevisae and Arabidopsis thaliana:

Useful model systems for the identification of molecular mechanisms involved in

resistance of plants to toxins. Eur. J. Plant Pathol. 108, 699-703 (2002).

38. Tommasini, R., et al. Differential expression of genes coding for ABC

transporters after treatment of Arabidopsis thaliana with xenobiotics. FEBS Lett.

411, 206-210 (1997).

39. Schaller, F., & Weiler, E.W. Molecular cloning and characterization of 12-

oxophytodienoate reductase, an enzyme of the octadecanoid signaling pathway

from Arabidopsis thaliana - Structural and functional relationship to yeast old

yellow enzyme. J. Biol. Chem. 272, 28066-28072 (1997).

40. Chang, S., Puryear, J., & Cairney, J. A simple and efficient method for isolating

RNA from pine trees. Plant Mol. Biol. Rep. 11, 113-116 (1993).

41. Paquette, S.M., Bak, S., & Feyereisen, R. Intron-exon organization and phylogeny

in a large superfamily, the paralogous cytochrome P450 genes of Arabidopsis

thaliana. DNA Cell Biol. 19, 307-17 (2000).

FIGURE LEGENDS

Figure 2.1. Chemical Structures of Nitroaromatic Explosives.

The nitroaromatic explosives of greatest concern for environmental contamination, TNT,

RDX, and HMX, are depicted.

Figure 2.2. Estimates of Transcriptome Size in Arabidopsis Roots.

Double-reciprocal plots of new tag discovery versus total tags sequenced are shown for

the TNT-treated (top panel) and control (bottom panel) root libraries. The estimated

transcriptome size was taken as the 1/y-intercept from the linear regression depicted in

each graph.

Table 2.1. SAGE tags induced at least 10-fold by exposure to TNT

Abundance

Increasea

tag sequence TNT Control

Control p-chanceb Locus Annotation

GTGAGTTTGA 30 0 >30.0 0 At2g18910

Possible signaling protein

CCAAATTCTG 55 2 27.5 0 At1g17170 Putative glutathione transferase

TAGCCAATTA 72 3 24.0 0 At5g11510 MYB-like protein

GGAGTTTGTA 23 1 23.0 0 At3g27740 Carbamoyl phosphate synthetase (small subunit)

ATGTTTCGCG 21 1 21.0 6.68E-06 At5g13750 Transporter-like protein

GGAGAAGTCC 20 1 20.0 1.67E-05 At1g60940 Putative serine/threonine-protein kinase

GCAATTCTAC 35 2 17.5 0 At5g03630 Monodehydroascorbate reductase-like protein

AGAAGTTTAT 17 1 17.0 8.67E-05 At5g63790 No apical meristem (NAM)-like protein

TGAGTTTCAA 17 1 17.0 8.67E-05 At4g01870

At5g60880

Unknown function

Putative protein

GGTTAGTCGA 15 1 15.0 3.13E-04 At1g05050 Putative transcription factor

AGAGAAAGTG 14 0 >14.0 5.34E-05 At3g28740 Putative cytochrome P450 (CYP81D11-A-TYPE)

CGGGGAAAAA 13 1 13.0 9.70E-04 ? Polyadenylation sequence?

AGTTGAGTTC 38 3 12.7 0 At5g45020 Putative glutathione transferase

GTGAAGTTTG 12 0 >12.0 1.90E-04 At3g45270 Putative protein

GGAAAAGGTG 23 2 11.5 2.34E-05 At3g21720 Putative isocitrate lyase

ACCAAAATTG 11 0 >11.0 5.10E-04 At2g29490 Putative glutathione S-transferase

CCACAGTTTT 30 3 10.0 0 At1g78080 Putative AP2 domain protein (RAP2-like transcription

factor, TINY)

AACGCAGAAA 10 0 >10.0 1.05E-03 At1g01550 Hypothetical protein

CAGGATGTGT 10 0 >10.0 1.05E-03 At1g05680 Putative indole-3-acetate beta-glucosyltransferase

GCACTCTTGA 10 0 >10.0 1.05E-03 At2g28570 Unknown protein (protein-tyrosine kinase motif)

CTTCTCTAGT 10 1 10.0 6.37E-03 At1g75270 Putative GSH-dependent dehydroascorbate reductase

CTTGTCCTCA 10 1 10.0 6.37E-03 At1g76680 12-oxophytodienoate reductase (OPR1)

aFor the calculation of expression ratios, a value of 1 was substituted where tag counts were zero. bP-chance values are averages of three Monte Carlo simulations *If tag is positioned at penultimate CATG site.

Table 2.2. Cytochrome P450 genes induced by exposure to TNT

Tag sequence Locus TNT Control TNT/control p-chance P450 Typea

AGAGAAAGTG At3g28740 14 0 >14.0 5.43E-05 CYP81D11 (putative) A

ACACCAAAGC At1g78490 11 2 5.5 1.17E-02 CYP708A3 (like protein) NA

GAAAAGATTT At2g30750 5 0 >5.0 3.10E-02 CYP71A12 A

GCTGAGAGAC At4g22690 5 0 >5.0 3.10E-02 CYP706A2 (like protein) A

ATGCGAAGCT At2g30490 41 9 4.6 3.34E-06 CYP73A5 (type enzyme) A

AAGCATCCGC At1g64950 4 0 >4.0 6.08E-02 CYP89A6 (hypothetical) A

aCytochrome P450 classification as described elsewhere41.

Table 2.3. TNT-induced SAGE tags representing potential detoxification pathway

components.

Abundance

increasea

tag sequence TNT Control

Control p-chancebLocus Annotation

TTGAGAAATT 9 1 9.0 1.06E-02 At5g54500 1,4-Benzoquinone reductase-like

ATTCTGAGAA 9 1 9.0 2.04E-03 At1g78340 Glutathione transferase-like protein

TGATGAGTTT 94 11 8.5 0 Metallothionein-like protein

TGGCGGATTA 40 5 8.0 0 At4g19880 Putative protein (related to glutathione S-tranferase)

AAGATCCAAG 8 1 8.0 1.96E-02 G-box binding bZIP transcription factor

TGCAAGTTAT 8 1 8.0 1.96E-02 At3g28480 Prolyl 4-hydroxylase, putative

TAGAATTCTC 7 1 7.0 3.50E-02 At4g03430 Putative pre-mRNA splicing factor

TGATTCAAAA 7 1 7.0 3.50E-02 At2g16680 Putative non-LTR retroelement reverse transcriptase

AAACTGTTTG 7 1 7.0 3.50E-02 At5g48930 Anthranilate N-benzoyltransferase

TCACTCCTAT 6 1 6.0 6.52E-02 At3g44300 Nitrilase 2

CAAATCAGTT 33 6 5.5 1.34E-05 At1g76930 Extensin

TCTCGAACCT 11 2 5.5 1.17E-02 Sucrose-UDP glucosyltransferase

GCTGTTTTTG 155 30 5.2 0 L-ascorbate peroxidase

TGTTTGGCTG 5 0 >5.0 3.10E-02 At3g58500 Phosphoprotein phosphatase 2A isoform

CCAATTAGTC 10 2 5.0 2.08E-02 ABC transporter - like protein

GAAACGCTCA 10 2 5.0 2.08E-02 Peptide methionine sulfoxide reductase-like protein

GTTTCGAGAT 47 10 4.7 0 Phospholipid hydroperoxide glutathione peroxidase

AAAACTCGGT 9 2 4.5 3.39E-02 Glutathione reductase, cytosolic

CTTGGTGCAA 11 3 3.7 3.07E-02 Citrate synthase

ACGAAGGTCG 9 3 3.0 7.08E-02 At3g44320 Nitrilase 3

TAACTTGTGC 20 7 2.9 1.06E-02 At1g51680 4-Coumarate:CoA ligase

GAAACTTAAA 31 11 2.8 1.45E-03 At4g30170 Peroxidase ATP8a

At3g09390

At2g18160

At5g20830

At1g07890

At3g53480

At5g07460

At4g11600

At3g24170

At2g44350

aFor the calculation of expression ratios, a value of 1 was substituted where tag counts were zero. bP-chance values are averages of three Monte Carlo simulations.

Table 2.4. SAGE tags repressed at least 14-fold by exposure to TNT

abundance

decreasea

Tag sequence Control TNT

Control/

TNT p-chanceb Locus Annotation

TTTTCTTACC 85 0 >85.0 0 At4g12550 Putative cell wall-plasma membrane linker protein

(AIR1A)

TATCCTTGTT 52 1 52.0 0 At2g01520 Major latex protein homolog-like

TTGTATGTTT 45 0 >45.0 0 At1g70850 Major latex protein homolog-like

TTTTCTTATC 37 0 >37.0 0 At4g12550 Putative cell wall-plasma membrane linker protein

(AIR1B)

GACCAACCAC 29 1 29.0 0 At2g36830 Putative aquaporin

TGTCTTAGCT 29 1 29.0 0 At1g09690 Putative 60S ribosomal protein L21

TTTATGCTTT 50 2 25.0 0 At4g22212 Expressed protein

TGTTTATTTT 24 1 24.0 0 At5g23710 Unknown protein

TTTTCTTCTT 23 1 23.0 3.34E-06 At3g54580 Extensin precursor-like protein

TTACAATAAC 21 1 21.0 0 At5g19510 Elongation factor 1B alpha-subunit

GGAACATATA 20 0 >20.0 0 At4g38740 Peptidylprolyl isomerase ROC1

GTGTTGTATG 20 0 >20.0 0 At1g14960 Putative major latex protein

TTGGTTATGT 20 0 >20.0 0 At5g56540 Arabinogalactan-protein (AGP14)

AACCCGGCCA 16 0 >16.0 2.34E-05 At4g17340 Membrane channel-like protein

TTTTTCTTTG 15 0 >15.0 2.00E-05 At4g12520 pEARLI 1-like protein

TCTTTGTCTT 15 1 15.0 1.87E-04 At5g10280 Putative transcription factor MYB92

AGTTTATCAC 14 0 >14.0 4.00E-05 At5g04750 F1F0-ATPase inhibitor-like protein

TTATCTCTCT 14 0 >14.0 4.00E-05 At1g04820 Tubulin alpha-2/alpha-4 chain

TTTTGCTATC 56 4 14.0 0 At5g26260 Putative protein

CAACATTGTA 42 3 14.0 0 At5g14200 3-isopropylmalate dehydrogenase

TTTCTGGTAA 14 1 14.0 4.90E-04 At4g14960 Tubulin alpha-6 chain (TUA6)

aFor the calculation of expression ratios, a value of 1 was substituted where tag counts were zero.

bP-chance values are averages of three Monte Carlo simulations.

Table 2.5. Comparison of transcript expression levels as determined by SAGE

and qPCR.

Control TNT Ratio

Amplicon/SAGE tag qPCR

Ct valuea

Ct value

count qPCRb

Cyt P450 35 0 29 14 28X increase

Ribosomal Protein 23 31 24 34 2X decrease

Aquaporin 25 29 29 1 11X decrease

aCt is a function of the starting number of templates present. bExpression ratios relative to control.

Figure 2.1.

NNO2O2N

NO2O2N

TNT (trinitrotoluene)

RDX (hexahydro-1,3,5-trinitro-

1,3,5-triazine)

HMX (octahydro-1,3,5,7-tetranitro-

1,3,5,7-tetrazocine)

Figure 2.2.

y = 1.0874x + 7E-05R2 = 0.9991

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

0 0.00005 0.0001 0.00015 0.0002 0.00025 0.0003 0.00035 0.0004

1/number of tags sequenced

Control

y = 1.2238x + 5E-05R2 = 0.997

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

0 0.00005 0.0001 0.00015 0.0002 0.00025 0.0003 0.00035 0.0004

1/number of tags sequenced

CHAPTER 3

EFFECTS OF THE EXPLOSIVE RDX ON TRANSCRIPT EXPRESSION IN

ARABIDOPSIS THALIANA SEEDLING ROOTS2

______________________________

2Ekman, D.R., Wolfe, N.L., and Dean, J.F.D. To be submitted to Environmental

Toxicology and Chemistry.

ABSTRACT

Arabidopsis thaliana root transcriptome responses to the munition 1,3,5-

hexahydro-1,3,5-triazine (RDX) were measured using serial analysis of gene expression

(SAGE). Sequencing of SAGE libraries from both control and RDX-exposed root tissue

to approximately 30,000 tags each revealed the induction of a number of general stress

response transcripts. Induced genes included ones encoding molecular chaperones and

transcription factors, as well as vacuolar proteins and peroxidases. Among the strongly

repressed genes were ribosomal proteins, a cyclophilin, a katanin, and a peroxidase.

Comparison of the transcriptional profile for the RDX response to that induced by

trinitrotoluene (TNT) exposure revealed significant differences. This suggests that

Arabidopsis employs drastically different mechanisms for coping with these two

compounds. With respect to the goal of engineering plants better capable of tolerating

and remediating explosives at contaminated sites, this study suggests that different genes

will be necessary to deal effectively with each type of explosive. (Note: the datasets

from these analyses are available at the ncbi Gene Expression Omnibus (GEO) at:

http://www.ncbi.nlm.nih.gov/geo/)

INTRODUCTION

Extensive contamination of a large number of munitions manufacturing and

disposal facilities throughout the U.S. and Europe over the past several decades by the

polynitramine, hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) (Figure 3.1), has created a

significant environmental problem. An important military explosive, RDX is a

compound of recognized toxicity (Cholakis et al., 1980). Often used as a rat poison,

RDX ingestion produces adverse effects in the central nervous system, gastro-intestinal

tract, and kidneys. Moreover, biological transformation of RDX can frequently serve to

increase its toxicity. Hexahydro-1,3,5-trinitroso-1,3,5-triazine, a degradation product of

RDX, has been used as an experimental tumorigen (Merck & Co., 1983; Sax, 1989).

As a cost-effective alternative to currently used methods for clean-up (i.e.,

excavation of the contaminated soil followed by incineration), phytoremediation, or the

use of plants to aid in the removal and/or degradation of environmental pollutants, is

being assessed for its potential to address RDX contamination (Best et al., 1997b; Best et

al., 1997a; Thompson et al., 1999; Bhadra et al., 2000). Unfortunately, although the

RDX in the surrounding environment was taken up rapidly by the plants in these studies,

most of it remained untransformed and ended up in the arial portions where it could be

introduced into the foodchain (Harvey et al., 1991). Therefore, plants capable of

tolerating high levels of RDX while simultaneously restricting the compound to the roots

and promoting its degradation there would be highly desirable. Unfortunately, virtually

nothing is known about the plant genes involved in RDX metabolism. Consequently, this

study was undertaken to identify key enzymes in the metabolism of this energetic

compound using serial analysis of gene expression (SAGE) (Velculescu et al., 1995;

Velculescu et al., 1997), a technique for transcriptional profiling. The SAGE technique

provides a quantitative snapshot of transcript levels in sampled tissues and has received

its widest application in the study of gene expression changes in human cancers (Riggins

and Strausberg, 2001). In this study we report changes in root transcript levels for liquid-

grown Arabidopsis seedling roots exposed to RDX and compare them with the transcript

profiles previously measured in similar tissues exposed to TNT.

MATERIALS AND METHODS

Plant Material, Growth Conditions, and Root Tissue Isolation

The following experimental descriptions were developed to conform as closely as

possible to the MIAME guidelines being promulgated for describing microarray

experiments (Brazma et al., 2001). A. thaliana ecotype Columbia seeds (WT-2, Lehle

Seeds, Round Rock, TX) were surface-sterilized by immersion in 70% ethanol for two

min, followed by immersion in a solution of 30% bleach, 0.2% triton X-100 for 40

minutes with gentle shaking. The sterilized seeds were then placed in sterile Murashige

and Skoog liquid medium prepared according to the manufacturer’s protocol (Invitrogen

Life Technologies, Carlsbad, CA). Plants were grown for 14 days at 250C under a 16-

hour photoperiod with constant shaking at 85 rpm in a growth chamber. RDX was

obtained from the U.S. Army Center for Environmental Health Research (Fort Detrick,

MD). Toxicity was assessed by adding RDX from a stock solution in DMSO to yield

final concentrations of 0, 5, 20, 40, 80, 100, and 150 mg/L in MS medium with cultures

run in triplicate. The cultures were returned to the growth chamber for five days, during

which time the seedlings in each flask were examined for signs of stress (leaf chlorosis

and necrosis). A final concentration of 150 mg/L RDX was judged to produce notable

stress in the plants without causing death. Root tissues for SAGE library construction

were isolated from seedlings grown 14 days in liquid MS medium and dosed with RDX

to a final concentration of 150 mg/L. Control tissues were isolated from seedlings grown

under the same conditions and dosed with an equivalent volume of DMSO. The

seedlings were grown in the presence of RDX or DMSO for 24 hours, after which they

were submerged briefly in dH2O to remove excess medium, and excised roots were

immediately frozen in liquid nitrogen. All root tissues were stored at –800C prior to RNA

extraction. The root tissues for each SAGE library were pooled from approximately 1400

seedlings recovered from 36 treatment flasks. Prior to seeding, the 36 flasks were

divided into three replicates of 12 with each group being started on different days.

RNA Isolation and cDNA Synthesis

of Chang et al. (1993). Poly(A) RNA was isolated from total RNA using Dynabeads

oligo-dT(25) magnetic beads (Dynal Biotech, Lake Success, NY) at a ratio of 0.25 mg of

total RNA per 250 µL of Dynabeads and following the manufacture’s instructions.

Double-stranded cDNA was synthesized from 5 mg of poly(A) RNA using the

Superscript Choice cDNA synthesis kit (Invitrogen Life Technologies) and following the

manufacturer’s protocol, except for the substitution of a 5’-biotin dT(18) primer in the

first-strand reaction.

SAGE library construction

SAGE libraries were constructed according to the SAGE Detailed Protocol,

Version 1.0c (Velculescu et al., 1997a), a brief description of which follows.

Biotinylated cDNAs from each tissue sample were bound to streptavidin-coated magnetic

beads and digested with NlaIII, a restriction enzyme recognizing the four-base sequence,

CATG (anchoring enzyme). DNA released by this digestion was washed away and the

beads, with the adherent 3’ ends of each cDNA, were split into two pools. Linkers

containing a binding site for BsmF1 (a Type-II restriction endonuclease – the tagging

enzyme), but different sites for PCR primers, were ligated to the NlaIII cleavage site at

the 5’ ends of the bead-bound cDNA fragments in each pool. Both pools of cDNAs were

digested with BsmF1 to release SAGE tags from the beads, after which the pools were

combined, and 102 bp linker-flanked ditags were formed by blunt-end ligation.

Following amplification of the ditags by PCR, the linkers were removed by NlaIII

digestion, and ditags were ligated to form concatemers. The concatemers were

subsequently size-fractionated, ligated into the pZero vector (Invitrogen Life

Technologies), cloned, and sequenced.

SAGE Data Analysis

Sequence files were compiled and analyzed using the SAGE Software, ver. 3.03,

provided by Dr. Kenneth Kinzler (Johns Hopkins University, Baltimore, MD). Tags

containing linker sequences and repeated ditags were excluded prior to analysis. Because

the library representing control roots was sequenced to a slightly greater extent (31,973

tags for the control versus 31,209 tags for the RDX treatment), values for the RDX

library tags were normalized prior to making comparisons of relative gene expression.

Ratios were used to compare the relative expression of tags between the two libraries,

and in instances where a particular tag was absent from a library, a value of 1 was

substituted to avoid division by zero. Using the SAGE software, Monte Carlo

simulations were performed to estimate the statistical significance of any differential

expression. The null hypothesis for these analyses was that the abundance, type, and

distribution of transcripts were the same in both libraries.

Gene Identification

To identify the genes from which the tags were derived, each 10-base tag plus the

4-base NlaIII recognition sequence was first compared against the Arabidopsis Gene

Initiative (AGI) database of model genes using the Patmatch analysis tool available on

The Arabidopsis Information Resource (TAIR) server (http://www.arabidopsis.org). If

the tag was found to match exactly the NlaIII site closest to the 3’ end of a model gene,

this identity was accepted for the tag. Tags that could not be found in the model gene

database were compared against all Arabidopsis sequences in GenBank using the same

Patmatch tool. Exact matches were annotated accordingly.

Quantitative PCR

Total RNA from Arabidopsis plants grown under conditions identical to those

used to generate RNA for the SAGE studies was used to independently verify the

expression of selected genes using quantitative RT-PCR. Messenger RNA isolated using

Dynal Oligo dT25 magnetic beads served as template for single-stranded cDNA synthesis

using Taqman Reverse Transcription Reagents (Applied Biosystems, Foster City, CA).

Real-time fluorescent detection of RT-PCR products was performed using a DNA Engine

Opticon System (MJ Research, Inc., Boston, MA) and Sybr Green PCR Master Mix

(Applied Biosystems). The following PCR primers were designed for this study using

Primer Express v. 1.0 (Applied Biosystems): nitrilase2 (At3g44300) forward, 5’-

GGACATTACTCGAGACCAGATGTTT -3’ and reverse, 5’-

TCGAAATGAATGTGACCGGTT -3’; peptidylprolyl isomerase (At2g39460) forward,

5’-CGAGATGGAGCTTTTCGCTG-3’ and reverse, 5’-

CCGGTACAGAGAGCACGGAA-3’; and putative HMG protein (At2g17560) forward,

5’-TGGTAAGGCTGCTGGAGCTAG -3’ and reverse, 5’-

GCTCTCAGCCTTAGCGACGT -3’. Amplification reactions were carried out

according to the manufacturer’s specifications as follows: two min at 500C followed by a

10 min activation of the enzyme at 950C, and 40 subsequent cycles consisting of 950C for

15 sec followed by 600C for 1 min. All amplification reactions were run using a dilution

series of cDNA from both control and RDX-treated tissues in the same PCR master mix.

Amplimers were checked for purity and size by gel electrophoresis to ensure that the

correct sequence was amplified. All samples were treated with DNAse prior to reverse

transcription using DNA-freeTM (Ambion, Inc., Austin, TX) to remove any contaminating

genomic DNA. Control reactions omitting reverse transcriptase were run for all samples

to ensure that genomic contamination did not contribute to the amplified products. Melt

curve analysis confirmed the absence of both primer-dimer and nonspecific product

formation.

RESULTS AND DISCUSSION

Toxicity of RDX to sterile hydroponic cultures of Arabidopsis thaliana

Sterile Arabidopsis seedlings were grown for 14 days in liquid Murashige and

Skoog medium before treatment with RDX at concentrations ranging from 5 to 150

mg/L. Visual signs of toxicity (mainly leaf chlorosis and necrosis) were not evident in

any of the cultures, except at the highest dosage (150 mg/L) after 72 hours of exposure.

While this concentration was in excess of the aqueous solubility of RDX (approximately

40 mg/L) it was nonetheless necessary to induce signs of toxicity in these plants. This

may be due to a requirement for accumulation of the compound within the plant tissues.

This would require a pool of RDX in the media that, although initially insoluble, would

enter into solution as portions of the soluble fraction were removed from the media by the

plants.

SAGE Libraries

SAGE libraries representing the transcripts expressed in Arabidopsis root tissues

grown in the presence or absence of RDX were sequenced until just over 30,000 tags

were characterized from each. The datasets from these SAGE library analyses are

available online in spreadsheet format

(http://www.arches.uga.edu/~jeffdean/SAGE/SAGEData.html). As shown in Table 3.1,

12,122 of the 31,209 tags characterized in the library for RDX-treated tissue represented

unique transcripts, and 7,909 of these were seen only once (singletons). Similarly, 12,719

unique tags were encountered amongst the 31,973 characterized in the control library,

with 8,322 of these representing singletons.

Differentially Expressed Transcripts

Significant differences in transcript expression were observed between the SAGE

libraries. In the presence of RDX, 135 tags displayed at least a five-fold increase in

expression over control. Moreover, a relatively large number of tags observed in the

RDX library (6672) were absent from the control library. Among the induced tags were

those representing transcripts from several different functional classes, including defense-

related proteins, transporters, transcription factors, and vacuolar proteins (Table 3.2).

The most highly induced tag corresponds to an NPR1-like protein previously shown to

respond to pathogen attack (Cao et al., 1994; Delaney et al., 1995; Glazebrook et al.,

1996; Shah et al., 1997; Cao et al., 1998; Chern et al., 2001; Friedrich et al., 2001). The

second most highly induced tag represented a DnaJ-like protein known to act as a

molecular chaperone or co-chaperone in association with Hsp70 (Kushnir et al., 1995;

Zhou et al., 1995; Kroczynska et al., 1996; Best et al., 1997b), which potentially signifies

stress-related problems with protein folding and/or stability in the cells exposed to RDX.

Also noteworthy is the induction of a vacuolar processing enzyme (γ-VPE) responsible

for the maturation of various cellular proteins, including a number that function in

cellular defense (Kinoshita et al., 1999). Cytochrome P450 enzymes are often utilized for

transformation of cellular toxins, either to make them more amenable to conjugation

(e.g., attachment of glutathione or six-carbon sugars) or to increase solubility. The

involvement of cytochrome P450 enzymes has been reported previously in the

degradation of RDX by two different strains of Rhodococcus sp., YH11 (Tekoah and

Abeliovich, 1999) and DN22 (Coleman and Duxbury, 1999). Interestingly, three

Arabidopsis cytochrome P450s were induced at least five-fold in the presence of RDX.

Also worth noting is the strong induction (12X) of a putative peroxidase transcript. This

may indicate oxidative stress in the plant roots or, alternatively, a mechanism for

oxidation of RDX. The fungus, Phanerochaete chrysosporium, is known to degrade the

nitroaromatic explosive, TNT, as well as a number of other organic pollutants using

oxidative enzymes such as peroxidases. Degradation of RDX by this fungus has been

reported to proceed by an oxidative mechanism, presumably catalyzed by lignin

peroxidases (Fernando and Aust, 1991).

Over 100 tags were repressed at least five-fold in response to RDX exposure

(Table 3.3). One of the tags showing the greatest repression corresponds to a protein

involved in cytokinesis (cell division), possibly indicating a reduced growth rate in the

presence of RDX. Other tags strongly repressed by RDX included a putative HMG (high

mobility group) protein. This class of proteins (five have been identified in Arabidopsis)

is thought to act as architectural elements in chromatin structure, and through their impact

on chromatin are suspected to influence such processes as transcription and

recombination (Crothers, 1993; Ner et al., 1994; Grosschedl, 1995; Grasser, 1998).

Decreased expression was also noted for a tag representing a glutaredoxin protein. This

protein is thought to function with ribonucleotide reductase in the production of

deoxyribonucleotides (Holmgren, 1989). Alternatively, rice glutaredoxin was found to

have dehydroascorbate reductase activity (Minakuchi et al., 1997), which along with the

potential for glutaredoxin to participate in general thiol reduction, suggests that oxidative

stress could result from repression of this gene. Interestingly, the tag for a putative

monodehydroascorbate reductase was also observed to decrease in abundance upon RDX

exposure.

Quantitative PCR

SAGE has previously been shown to provide an accurate reflection of gene

expression levels for medium- and high-abundance transcripts, but the induction or

repression of selected Arabidopsis transcripts in response to RDX exposure was

independently verified in this study using real-time quantitative PCR. Genes that SAGE

analysis suggested were induced (At3g44300; nitrilase 2), repressed (At2g17560; HMG

protein), or remained unaffected (At2g39460; peptidyl prolyl isomerase) by RDX

exposure were tested. As shown in Table 3.4, the quantitative PCR data were in general

agreement with the SAGE data. Transcripts measured as induced or repressed using

SAGE were observed to follow similar trends when quantitated using qPCR. However,

the transcript unaffected by the presence of RDX as measured using SAGE displayed a 3-

fold decrease in abundance according to qPCR. The relatively high expression levels for

this transcript indicated by SAGE (tag counts of around 40 in both libraries) would

suggest that a sampling effect is not responsible for this contradiction. It is more likely

that transcripts for other peptidyl-prolyl isomerase enzymes with similar sequences may

have been detected by the qPCR technique. In fact, the SAGE data revealed repression

of three other transcripts for variants (At4g38740, At1g265502, At2g47320) of this

enzyme.

Differences In Transcript Expression between TNT-exposed and RDX-exposed

Arabidopsis Roots

A previous study of gene expression changes induced in Arabidopsis by the

presence of another toxic explosive, 2,4,6-trinitrotoluene (TNT), identified several highly

induced transcripts (Ekman et al., 2003). A comparison with the RDX treatment data

reveals significant differences in the responses to these two compounds (Table 3.5). For

example, the tags representing different cytochrome P450 enzyme transcripts vary

considerably between the treatments. Whereas a putative CYP81D11 transcript was

induced at least 14-fold (14 to 0) in the TNT SAGE library, it was not detected in RDX-

treated tissues. Also absent from the RDX-treated tissue was a transcript encoding a

glutathione S-transferase, the most highly induced transcript (27X) in the TNT library.

Tags representing oxidative stress enzymes, such as monodehydroascorbate reductase,

were low in the RDX library while their expression was highly induced in the TNT-

treated roots (data not shown). Thus, while metabolism of TNT in plants probably

requires a multiphase oxidative mechanism, as was described previously (Ekman et. al.

2003) and as has been observed in plants for a variety of other compounds, this may not

be the case for RDX. This will be a major consideration in the development of plants for

remediation of sites contaminated with RDX. Furthermore, if plants are to be able to

remediate both of these explosives simultaneously (TNT and RDX are often found

together at polluted sites) it may require the introduction of suites of genes specific for

each compound.

Analysis of transcripts expressed commonly among all three libraries (control,

TNT, and RDX) and those unique to each treatment gives an indication of the specificity

of gene expression induced by each compound. As indicated in Figure 3.2, it appears as

though a number of genes (3,572) cannot be dispensed with despite adverse conditions.

Most likely, the majority of these are housekeeping genes or those specifically employed

for growth under the culture conditions used. On the other hand, a significant number of

genes were specific to each library. The largest number of genes expressed specifically

for any given library was that for the control library (2,185). This potentially indicates

the reduced expression of genes that are not completely necessary for survival in the

munition-treated plants (1,994 for TNT and 1,745 for RDX). This sacrifice may be a

result of the increased demand for higher expression levels of specific genes utilized in

responding to toxicity. It should be noted that these numbers are derived from a depth of

sequencing of around 30,000 tags in each library. Further sequencing will likely alter

these numbers.

CONCLUSIONS

From these analyses it appears as though the mechanism for RDX metabolism in

Arabidopsis is vastly different than for TNT. This is somewhat expected from the results

of previous studies analyzing the fate of RDX in plants (Harvey et al., 1991; Best et al.,

1997b; Burken and Schnoor, 1998; Thompson et al., 1999). Identification of induced

cytochrome P450 transcripts suggests a potential route to the tranformation of RDX,

although the functions of these enzymes may not be directly related to RDX metabolism,

but simply reflect a response to the general stress imposed by toxicity.

Future studies must include analyses of transcriptome responses in leaf and stem

tissues. As these tissues are commonly sites of RDX accumulation in plants, discovery of

differential gene expression in these locations is crucial to achieving a full understanding

of this compound’s metabolism in plants. Such studies are underway using high-density

DNA microarrays to measure expression changes in these tissues and to observe the

effects of longer periods of exposure and different concentrations of RDX. Eventually,

transcriptome analyses of soil-grown plants exposed to RDX will be necessary to

determine responses likely to be found under field conditions.

FIGURES LEGENDS Figure 3.1. Explosive compounds of particular concern as environmental contaminants.

TNT and RDX are highly recalcitrant, toxic compounds found extensively in soil, as well

as in ground and surface water, at or near many military installations.

Figure 3.2. Diagrammatic representation of the tags unique to each library (control,

TNT, or RDX) as well as those shared among subsets of the libraries.

TABLES Table 3.1. SAGE library statistics. Frequency distributiona

≥20 19 to 5 4 to 2 =1 Totals

RDX library

Unique tags 153 (1.3)b 1,021 (8.4) 3,039 (25.1) 7,909 (65.2) 12,122

Tags sequenced 7,063 (22.6) 8,498 (27.2) 7,739 (24.8) 7,909 (25.4) 31,209

Control library

Unique tags 178 (1.4) 1,005 (7.9) 3,214 (25.3) 8,322 (65.4) 12,719

Tags sequenced 7,276 (22.8) 8,219 (25.7) 8,156 (25.5) 8,322 (26) 31,973

aFrequency distributions were calculated based on the total number of unique or sequenced tags in each library shown in the totals column. bThe percentages of tags in each frequency group are shown parenthetically

Table 3.2. SAGE tags induced at least 7-fold by exposure to RDX

Abundance Fold

Increasea

tag sequence Control RDX RDX/Control p-chanceb Locus Annotation

GTGGTAACGG 1 30 30.0 0.00 At4g26120 NPR1 like protein TGCTTACCGT 1 14 14.0 0.00 At4g36040 DnaJ-like protein GGATAACATC 1 14 14.0 0.00 At4g32940 gamma-VPE (vacuolar processing

enzyme) GGTTAGTCGA 1 13 13.0 0.00 At1g05050 putative transcription factor CGCTGACATA 1 12 12.0 0.00 At1g49570 peroxidase, putative TTCAAGTCCA 1 11 11.0 0.00 At5g66120 3-dehydroquinate synthase, putative GCCGTTCTTA 3 32 10.7 0.00 At3g42050 Vacuolar H+-ATPase subunit H (VHA-H) TCTTCTCGAA 1 10 10.0 0.01 At4g38920 H+-transporting ATPase 16K chain P2,

vacuolar GTGATGCTCT 1 10 10.0 0.01 At5g10300 alpha-hydroxynitrile lyase-like protein TCCCCTATTA 1 9 9.0 0.01 ******** no matches in genome GGAGACAGTG 1 9 9.0 0.01 At4g18930 putative protein GATGTCTGGC 1 9 9.0 0.01 At1g35516 myb family transcription factor TTTGGAGGAG 2 17 8.5 0.00 At4g19200 similarity to glycine/proline-rich protein

GPRP TTCCTATTCT 2 17 8.5 0.00 At2g45960 aquaporin (plasma membrane intrinsic

protein 1B) TTTCTGGATA 1 8 8.0 0.02 At5g20150 ids4-like protein TAATGTAATG 0 8 ≥8.0 0.00 At5g04590 sulfite reductase CCTGGAATCA 1 8 8.0 0.02 ******** no matches in genome ATCCTTGTCT 1 8 8.0 0.02 At2g37110 unknown protein

AGAAAGCAGG 1 8 8.0 0.02 At5g35200 putative protein TTTTTTTTCT 0 7 ≥7.0 0.01 2 HITS AT1G33190, auxin-response factor

(fragment) and AT3G11650.1, hairpin-induced protein, putative (HIN1)

Table 3.3. SAGE tags repressed at least 8-fold by exposure to RDX

Tag Abundance

Fold Increasea

tag sequence Control RDX Control/RDX p-chanceb Locus Annotation

GTTCTGCAAA 13 1 13.0 0.00 hits on chromosomes 2 & 5 TTTAAAAAAA 11 1 11.0 0.00 At5g27030 putative protein; cytokinesis (cell

division) TCTGAAAGAG 11 1 11.0 0.00 At3g05590 putative 60S ribosomal protein L18 TTTCACACTT 10 1 10.0 0.01 At4g30170 peroxidase ATP8a TTGGTGTTTC 10 1 10.0 0.01 At5g63030 glutaredoxin-like protein GCCCTGCGAT 10 1 10.0 0.01 one hit on chromosome 4 AAAGGCGGCG 10 1 10.0 0.01 At2g17560 putative HMG protein GGACAGATTC 9 0 ≥9.0 0.00 ribosomal protein S20 GGAAATGAAC 9 0 ≥9.0 0.00 genome hits chromosome 1, 2, 4, 5 GATCGACCAA 9 1 9.0 0.01 2 hits At1g20310; hypothetical protein

At1g74030; putative enolase ATCATCATCA 9 1 9.0 0.01 ******* penultimate CATG At4g27840 ???

(numerous genome hits in both orientations --SSRs)

TAATTATGTA 8 0 ≥8.0 0.00 At4g39675 Expressed protein

TAAGAAATCT 8 1 8.0 0.02 ras-related small GTP-binding protein RAB1c

TAAAAGCTTT 8 1 8.0 0.02 At2g47320 putative peptidyl-prolyl cis-trans isomerase

GTCTTAATGA 8 0 ≥8.0 0.00 genome hits chromosome 1 & 2 GGAAATAAAA 8 1 8.0 0.02 At1g27030 expressed protein GACGCTAGCG 8 1 8.0 0.02 At2g10410 pseudogene ATGTGTTGGT 8 0 ≥8.0 0.00 At2g34560 putative katanin ATCAAAAAAA 8 0 ≥8.0 0.00 one hit on chromosome 3 (1 ESTs) AGGAAGGAAG 8 1 8.0 0.02 At5g03660 putative protein

At5g62300

At4g17530

Table 3.4. Comparison of transcript expression levels as determined by SAGE and qPCR.

Control RDX Ratio

Amplicon/SAGE tag qPCR

Ct valuea

Ct value

count qPCR b

nitrilase 2 (At3g44300) 18 1 17 7 2X increase

peptidyl prolyl isomerase (At2g39460) 15 44 17 42 3X decrease

HMG protein (At2g17560) 20 10 21 1 2X decrease

aCt is a function of the starting number of templates present. bExpression ratios relative to control.

Table 3.5. Differential expression of selected genes in Arabidopsis roots dependent upon the type of explosive (either TNT or RDX) used for exposure.

Tag Abundance

Fold Increase

for RDXa

tag sequence Control TNT RDX RDX/TNT Locus Annotation

TNT-specific cytochrome P450s

AGAGAAAGTG 0 14 0 0.07 At3g28740 cytochrome P450, putative; CYP81D11 GCTGAGAGAC 0 5 0 0.20 At4g22690 cytochrome P450 like protein; CYP706A2 GAAAAGATTT 0 5 0 0.20 At2g30750 cytochrome P450; CYP71A12

RDX-specific cytochrome P450s

TATGCCGCCC 1 0 5 5.00 At1g16400 cytochrome P450-like protein; CYP79F2

cytochrome P450 common to both

ACCTAACTGA 0 5 5 1.00 At4g13310 cytochrome p450-like protein; CYP71A20 ACACCAAAGC 2 11 3 0.27 At1g78490 cytochrome P450-like protein; CYP708A3 AATAAAACTT 2 9 1 0.11 At2g24180 cytochrome P450, putative; CYP71B6 AAGCATCCGC 0 4 1 0.25 At1g64950 cytochrome P450, hypothetical protein; CYP89A6 ATTGCGCGTG 1 1 5 5.00 At3g20940 cytochrome P450, putative; CYP705A30

aFor the calculation of expression ratios, a value of 1 was substituted where tag counts were zero.

Figure 3.1

NNO2O2N

NO2O2N

NO2 RDX

(hexahydro-1,3,5-trinitro-1,3,5-triazine)TNT

(trinitrotoluene)

Figure 3.2

TNT RDX

Control

REFERENCES Best, E.H.P., Sprecher, S.L., Fredrickson, H.L., Zappi, M.E., and Larson, S.L.

(1997a). Screening submersed plant species for phytoremediation of explosives

contaminated groundwater from the Milan Army Ammunition Plant, Milan,

Tennessee. Technical Report EL-97-24. U.S. Army Engineer Waterways

Experiment Station, Vicksburg, MS.

Best, E.H.P., Miller, J.L., Fredrickson, H.L., Larson, S.L., Zappi, M.E., and

Streckfuss, T.H. (1997b). Screening of aquatic and wetland plant species for the

Ammunition Plant. Ann. N.Y. Acad. Sci. 829, 179-194.

Bhadra, R., Williams, R., Barman, S., Stone, M.B., Hughes, J.B., and Shanks, J.V.

(2000). Fate of RDX and HMX in axenic plant roots. Chemosphere 44, 1259-

Burken, J.G., and Schnoor, J.L. (1998). Predictive relationships for the uptake of

organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32, 3379-

Cao, H., Li, X., and Dong, X. (1998). Generation of broad-spectrum disease resistance

by overexpression of an essential regulatory gene in systemic acquired resistance.

Proc. Natl. Acad. Sci. USA 95, 6531-6536.

Cao, H., Bowling, S.A., Gordon, S., and Dong, X. (1994). Characterization of an

Arabidopsis mutant that is nonresponsive to inducers of systemic acquired

resistance. Plant Cell 6, 1583-1592.

Chern, M.S., Fitzgerald, H.A., Yadav, R.C., Canlas, P.E., Dong, X., and Ronald,

P.C. (2001). Evidence for a disease-resistance pathway in rice similar to the

NPR1-mediated signaling pathway in Arabidopsis. Plant J. 27, 101-113.

Cholakis, J.M., Wong, L.C.K., Van Goethem, D.L., Minor, J., Short, R., Sprinz, H.,

and Ellis III, H.V. (1980). Mammalian toxicological evaluation of RDX.

Contract No. DAMD17-78-C-8027. U.S. Army Medical Research and

Development Command, Fort Detrick, Frederick, MD.

Coleman, N., and Duxbury, T. (1999). Biodegradation of RDX by Rhodococcus sp.

strain DN22. Second Int. Symp. Biodegradation of Nitroaromatic Compounds

and Explosives. Leesburg, VA, Sept. 8-9, Abstract p. 13.

Crothers, D.M. (1993). Architectural elements in nucleoprotein complexes. Curr. Biol.

3, 675-676.

Delaney, T.P., Friedrich, L., and Ryals, J.A. (1995). Arabidopsis signal transduction

mutant defective in chemically and biologically induced disease resistance. Proc.

Natl. Acad. Sci. USA 92, 6602-6606.

Fernando, T., and Aust, S.D. (1991). Biodegradation of munition waste, TNT (2,4,6-

trinitrotoluene), and RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine) by

Phanerochaete chrysosporium. ACS Symposium Series. Ind. Eng. Chem. 486,

214-232.

Friedrich, L., Lawton, K., Dietrich, R., Willits, M., Cade, R., and Ryals, J. (2001).

NIM1 overexpression in Arabidopsis potentiates plant disease resistance and

results in enhanced effectiveness of fungicides. Mol. Plant-Microbe Interact. 14,

1114-1124.

Glazebrook, J., Rogers, E.E., and Ausubel, F.M. (1996). Isolation of Arabidopsis

mutants with enhanced disease susceptibility by direct screening. Genetics 143,

973-982.

Grasser, K.D. (1998). HMG1 and HU proteins: architectural elements in plant

chromatin. Trends Plant Sci. 3, 260-265.

Grosschedl, R. (1995). Higher-order nucleoprotein complexes in transcription:

analogies with site-specific recombination. Curr. Opin. Cell Biol. 7, 362-370.

Harvey, S.D., Fellows, R.J., Cataldo, D.A., and Bean, R.M. (1991). Fate of the

explosive hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) in soil and

bioaccumulation in bush bean hydroponic plants. Environ. Toxicol. Chem. 10,

845-855.

Holmgren, A. (1989). Thioredoxin and glutaredoxin systems. J. Biol. Chem. 264, 13963-

13966.

Kinoshita, T., Yamada, K., Hiraiwa, N., Kondo, M., Nishimura, M., and Hara-

Nishimura, I. (1999). Vacuolar processing enzyme is up-regulated in the lytic

vauoles of vegetative tissues during senescence and under various stressed

conditions. The Plant J. 19, 43-53.

Kroczynska, B., Zhou, R.G., Hayman, G.T., and Miernyk, J.A. (1996). AtJ1, a

mitochondrial homologue of the Escherichia coli DnaJ protein. Plant Mol. Biol.

31, 619-629.

Kushnir, S., Babiychuk, E., Kampfenkel, K., Bellesboix, E., Vanmontagu, M., and

Inze, D. (1995). Characterization of Arabidopsis thaliana cDNAs that render

yeasts tolerant toward the thio-oxidizing drug diamide. Proc. Natl. Acad. Sci.

USA 92, 10580-10584.

Merck & Co. (1983). The Merck Index, 10th ed. (Rahway, NJ).

Minakuchi, S.S., Higaki, N., Sato, K., Ohtsuki, K., Kurata, A., Yoshikawa, H.,

Kotaru, M., Masumura, T., Ichihara, K., and Tanaka, K. (1997). Purification

and characterization of glutaredoxin (thioltransferase) from rice (Oryza sativa L.).

J Biochem (Tokyo) 121, 842-848.

Ner, S.S., Travers, A.A., and Churchill, M.E.A. (1994). Harnessing the writhe: a role

for DNA chaperones in nucleoprotein complex formation. Trends Biochem. Sci.

19, 185-187.

Riggins, G.J., and Strausberg, R.L. (2001). Genome and genetic resources from the

Cancer Genome Anatomy Project. Hum. Mol. Genet. 10, 663-667.

Sax, N.I., and Lewis, R.J. (1989). Dangerous Properties of Industrial Materials. (New

York, New York: VanNostrand Reinhold).

Shah, J., Tsui, F., and Klessig, D.F. (1997). Characterization of a salicylic acid-

insensitive mutant (sai1) of Arabidopsis thaliana identified in a selective screen

utilizing the SA-inducible expression of the tms2 gene. Mol. Plant-Microbe

Interact. 10, 69-78.

Tekoah, Y., and Abeliovich, N.A. (1999). Participation of cytochrome P-450 in the

biodegradation of RDX by a Rhodococcus strain. Second Int. Symp.

Biodegradation of Nitroaromatic Compounds and Explosives. Leesburg, VA,

Sept. 8-9, Abstract p. 7.

Thompson, P.L., Ramer, L., and Schnoor, J.L. (1999). Hexahydro-1,3,5-trinitro-1,3,5-

triazine translocation in poplar trees. Environ. Toxicol. Chem. 18, 279-284.

Velculescu, V.E., Zhang, L., Vogelstein, B., and Kinzler, K.W. (1995). Serial analysis

of gene expression. Science 270, 484-487.

Velculescu, V.E., Zhang, L., Zhou, W., Vogelstein, J., Basrai, M.A., Bassett, D.E.,

Hieter, P., Vogelstein, B., and Kinzler, K.W. (1997). Characterization of the

yeast transcriptome. Cell 88, 243-251.

Zhou, R.G., Kroczynska, B., Hayman, G.T., and Miernyk, J.A. (1995). AtJ2, an

Arabidopsis homolog of Escherichia coli DnaJ. Plant Physiol. 108, 821-822.

CHAPTER 4

MICROARRAY ANALYSIS OF ARABIDOPSIS TRANSCRIPTOME CHANGES

INDUCED BY TNT

INTRODUCTION

Microarrays

Measurement of changes in gene expression can be an effective method for

studying numerous biological phenomena, including disease states, morphogenesis, and

toxicological effects (Deyholos and Galbraith, 2001). Although many methods exist to

aid researchers in these analyses (see Chapter 1), high-density DNA microarrays provide

for the parallel quantification of expression levels for large numbers of genes at a

relatively low cost. Analyses on this scale provide for the characterization not only of

individual genes, but also the observation of multigenic patterns of expression (Brown

and Botstein, 1999), which can aid in the elucidation of gene function.

The two major types of microarrays used for transcript expression profiling are

DNA microarrays, which utilize larger DNA elements (400-2000 bp) that are amplified

from cloned sequences using PCR, and oligonucleotide microarrays that employ smaller

DNA elements (20-90 bp). These may be synthesized prior to spotting onto the glass

support or built directly on the support using a photolithographic process (McGall et al.,

1996; Lipshutz et al., 1999). Although the longer length of PCR products has potential

for greater sensitivity, such microarrays can suffer from nonspecific hybridization due to

similarities between gene family members or protein domains. However, specificity can

be lost when using oligonucleotides at the low end of the size range, forcing the need to

strike a balance between the two extremes (Lashkari et al., 1997). For this reason longer

oligonucleotides (60-90 bp) are gaining in popularity because of their balance between

sensitivity and specificity (Figure 4.1).

Figure 4.1. The ability of longer oligonucleotide DNA elements (specifically 70-mers) to

strike a balance between both specificity and sensitivity as compared to both shorter

oligonucleotides and longer oligonucleotides/cDNA fragments. Diagram from the

Qiagen website: (http://www.qiagen.com).

Arabidopsis Oligonucleotide Microarrays

To assess changes in the Arabidopsis transcriptome induced by energetic

compounds, previous studies utilized serial analysis of gene expression (SAGE).

However, to fully appreciate plant responses to these compounds, changes must be

measured under a variety of conditions and within different plant tissues. Whereas costs

are prohibitive to use SAGE for such studies, microarray technology is better suited for

this type of analysis. Consequently, the use of high-density oligonucleotide microarrays

representing 26,090 genes (well characterized genes and predicted open reading frames)

from the Qiagen-Operon Arabidopsis Genome Oligo Set Version 1.0 (Qiagen Operon,

Alameda, CA) was initiated. Preliminary analyses using conditions identical to those

used for the TNT SAGE studies (15 mg/L TNT) have been conducted. These initial

studies have revealed striking similarities in the identification of differential expression

by both techniques. However, the limitations characteristic of each technique were also

revealed confirming the need to avoid using either one in isolation.

MATERIALS AND METHODS

Plant Material, Growth Conditions, and Root Tissue Isolation

The following experimental descriptions were developed to conform as closely as

possible to the MIAME guidelines for describing microarray experiments (Brazma et al.

2001). A. thaliana ecotype Columbia seeds (WT-2, Lehle Seeds, Round Rock, TX) were

surface-sterilized and placed in sterile Murashige and Skoog liquid medium prepared

according to the manufacturer’s protocol (Invitrogen Life Technologies, Carlsbad, CA).

Plants were grown for 14 days at 250C under a 16-hr photoperiod with constant shaking

at 85 rpm in a growth chamber. TNT was obtained from the U.S. Army Center for

Environmental Health Research (Fort Detrick, MD). Root tissues for microarray

analyses were isolated from seedlings grown 14 days in liquid MS medium and dosed

with TNT dissolved in DMSO to a final concentration of 15 mg/L. Control tissues were

isolated from seedlings grown under the same conditions and dosed with an equivalent

volume of DMSO. The seedlings were grown in the presence of TNT or DMSO for 24

hr, after which they were submerged briefly in dH2O to remove excess medium, and

excised roots were immediately frozen in liquid nitrogen. All root tissues were stored at

–800C prior to RNA extraction. The root tissues for each condition (i.e., TNT-dosed or

control) were pooled from approximately 1400 seedlings recovered from 20 treatment

flasks grown in parallel.

RNA Isolation

of Chang et al. (1993). Poly(A) RNA was isolated from total RNA using Dynabeads

oligo-dT(25) magnetic beads (Dynal Biotech, Lake Success, NY) at a ratio of 0.25 mg of

total RNA per 250 µL of Dynabeads and following the manufacture’s instructions.

Labeling cDNA for Oligonucleotide Microarray Hybridization

The following protocol was adapted from a Galbraith lab protocol

(http://ag.arizona.edu/microarray/protocol1.doc). Messenger RNA isolated from each

condition under study was labeled for microarray analysis as follows: 2 µg of poly A+

mRNA (in a volume of 50 µL) was combined with a mixture of dNTPs (3.0 µL of a

mixture of 10 mM dATP, dCTP, dGTP (350 µM final), 2 mM dTTP (70 µM final))

(Sigma-Aldrich, St. Louis, MO), either Cy5-dUTP or Cy3-dUTP (2.0 µL) (24 µM final)

(Amersham Biosciences, Piscataway, NJ), and random 15-mer primers (2.0 µL of 0.5

µg/µL solution for a 10 ng/µl final concentration) (Qiagen Inc., Valencia, CA) in a total

volume of 57 µL. The solution was heated for five min at 650C, after which 5X first

strand buffer (17 µL) (BD Biosciences, Clontech, Palo Alto, CA), 0.1 M DTT (8.0 µL),

RNAse inhibitor (1.0 µL) (Invitrogen Life Technologies, Carlsbad, CA), and reverse

transcriptase (1.0 µL) (BD Biosciences Clontech, Palo Alto, CA) were added to yield a

total volume of 84 µL. The solution was then heated for an additional two hr at 420C,

after which the reaction was terminated by the addition of EDTA (5 µL of 0.5 M) and the

RNA was removed using NaOH (5 µL of 1 M). The solution was then heated to 650C for

a period of 10 min, followed by the addition of Tris-HCl (pH 8.0) (25 µL of 1 M) and TE

(100 µL), prior to storage in the dark at –200C until use. Prior to hybridization, all

labeled targets were purified using Microcon YM-30 nitrocellulose spin columns

(Millipore Corp., Billerica, MA).

Preparation of the Oligonucleotide Microarrays

(http://ag.arizona.edu/microarray/protocol1.doc). Printed in the Galbraith laboratory at

The University of Arizona, the high-density arrays used in these experiments consist of

70-mer oligonucleotides printed on aminosilane-coated slides (Telechem) arrayed in a

format to accommodate the presence of 27,216 elements.

Spotted oligonucleotides are not immobilized on the glass slides prior to

distribution, and this must be done prior to their use in experiments. To accomplish this,

the microarrays were held over a 500C water bath for 10 seconds to rehydrate, followed

by snap-drying on a 650C heating block for 5 seconds. For each array this process was

repeated four times. Following rehydration, the arrays were crosslinked using a

Stratalinker UV Crosslinker (Stratagene, La Jolla, CA) set at 130 mJ. After crosslinking

the slides were washed in a solution of 1% SDS for five minutes at room temperature

with gentle shaking. SDS was removed from the array by gentle shaking in 100%

ethanol for 30 seconds. The arrays were then dried by centrifugation at 1000 rpm for two

minutes. Slides were stored desiccated in the dark until use.

Hybridization of Microarrays

(http://ag.arizona.edu/microarray/protocol1.doc). Prior to hybridization, 20X SSC (12.0

µL), Liquid Block (7.2 µL) (Amersham Biosciences, Piscataway, NJ), 2% SDS (4.8 µL),

and dH20 to a final volume of 120 µL was added to the purified sample. This solution

was heated over boiling water for 2 min and rapidly transferred to ice. Hybridizations

were carried out using a Genomic Solutions Automated Hybridization Chamber

programmed as follows: O-ring stuff (750C for 2 min), probe introduction (650C),

hybridization (550C for 8 hr with agitation), 2X SSC, 0.5% SDS wash (550C; flow for 16

sec; hold for 5 min), 0.5X SSC wash (250C; flow for 16 sec; hold for 5 min), 0.05X SSC

wash (250C; flow for 16 sec; hold for 5 min). Following hybridization, slides were rinsed

with gentle shaking in dH2O for 10 sec and spun (1,000 rpm for 5 min) or blown dry

(filtered air).

Scanning of Microarrays

Microarrays were scanned using a Scanarray 5000 spectrometer (GSI Lumonics,

Billerica, MA) with appropriate laser and filter settings. Data was extracted using

Quantarray software (Applied Biosystems, Foster City, CA) selecting background

subtraction and quantification via the fixed circle method. Data were normalized by

measuring the total intensity of each array on the assumption that the majority of spots

represented genes whose expression was unaltered by treatment. Spots whose intensities

were altered by local effects (e.g., dust, residual wash buffer, etc.) were flagged and

removed from analyses.

RESULTS AND DISCUSSION

Differentially expressed transcripts

Microarray analysis identified numerous Arabidopsis transcripts that were

differentially expressed in response to TNT (Table 4.1). Striking were the increased

levels of several glutathione transferase and glutathione peroxidase transcripts (Table

4.1). Transcripts for glucosyltransferase enzymes were also highly induced, suggesting

an up-regulation of conjugating enzymes by TNT. Other highly induced transcripts

included those for transcription factors, oxidative stress response enzymes, quinone

oxidoreductases, and cytochrome P450s. The most highly induced transcript was that for

nitrilase 2, an enzyme thought to be responsible for the conversion of indoleacetonitrile to

the plant growth factor indole-3-acetic acid (auxin) (Normanly et al., 1997).

Several transcripts were also observed to decrease upon TNT treatment (Table

4.2). A transcript for a pEARLI-1-like protein showed the greatest repression in response

to TNT. Repression of this transcript along with those for a cell wall plasma membrane-

disconnecting CLCT protein (AIR1A), extensins, and an arabinogalactan protein suggests

that TNT has a strong impact on plant cell wall formation. Repression was also observed

for transcripts related to vacuolar functions, such as gamma tonoplast intrinsic proteins

and major latex proteins.

Comparison of Data Derived From Microarrays and SAGE

Although the reliability of data from microarray analyses hinges on the use of

replicates for statistical power, it is interesting to compare these preliminary microarray

data to the results obtained using SAGE to analyze Arabidopsis seedling roots treated in

an identical fashion (exposure to 15mg/L TNT for 24 hr). As the fundamental processes

underlying these techniques differ completely (i.e., DNA sequencing vs. differential

hybridization), the identification of the same differentially expressed transcripts with both

techniques strongly suggests that these gene products have biological relevance for the

plant’s response to TNT.

In general, genes that showed strong induction by TNT as measured using SAGE,

were strongly induced in the microarray analyses (Table 4.3). For example, a

glutathione-S-transferase (GST) transcript (At1g17170) was the tag most highly induced

in the SAGE studies. Transcripts for this enzyme were the third most highly induced in

the microarray analyses, and transcripts for other GSTs comprised six of the top 20 most

highly induced by TNT. Similarly, the most highly repressed transcripts were also

comparable for the two techniques (Table 4.4).

This effect (general agreement between the two techniques for the most highly

differentially expressed transcripts) was also observed in another study in which the

SAGE technique was compared to GeneChip technology (Affymetrix, Inc., CA) for

measuring changes in gene expression in human blood monocytes and granulocyte-

macrophage colony-stimulating factor induced macrophages (Ishii et al., 2000).

Furthermore, it is known that the validity of microarray data increases with the number of

replicates used (Lee et al., 2000). Thus, agreement between the techniques is likely to

increase as the experiment is repeated.

While many of the other high ranking differentially expressed transcripts were not

identical matches between SAGE and microarrays, many were similar in function. This

may result from the use of a different pool of plants for the two studies. However, it may

also be that cross-hybridization occurred in regions where family members possess

sequence similarities. This may have been the cause of the apparent disagreement

between the techniques regarding the most highly induced GST. While SAGE indicated

the GST encoded by gene At1g17170 as being the most highly induced, the microarray

results pointed to gene At1g17180, a family member located immediately downstream

from At1g17170 (Table 4.3).

General Comparisons of SAGE and Microarrays

This work also allowed for head-to-head comparison of two fundamentally

different transcriptional profiling methods: SAGE and high-density microarrays. While

both of these methods provide broad analyses of the transcriptome, each has distinct

strengths and weaknesses. In terms of time, microarray analysis is far superior.

Hybridization, scanning, and data extraction can all be achieved within days. Since

hybridizations can be run in parallel, the experimental replicates required for statistically

significant analyses can be performed simultaneously (Lee et al., 2000). In contrast, the

construction of SAGE libraries typically consumes a period of two weeks for each

library, and sequencing can consume at least another two weeks depending on the

required depth of analysis and the availability of high-throughput sequencing equipment.

This time factor is a major consideration in the design of experiments for measurement of

transcript levels under several different conditions. Furthermore, these time estimates

only include the work required to obtain measures of transcript levels and do not take into

account the time required for bioinformatic analyses and data mining. In terms of cost,

microarrays also have the advantage, provided the required equipment (e.g., hybridization

chambers, scanner, spotting robot, etc.) is available. These cost differentials hinge

predominantly on the lack of need for extensive DNA sequencing and the costly reagents

it consumes. Finally, the requirement for polyadenylated transcripts by SAGE precludes

its use for analysis of prokaryotes, while microarrays are not limited by this requirement.

Despite significant advantages, unlike SAGE, microarrays cannot aid in gene discovery.

Moreover, as SAGE does not require extensive genome analysis, changes in

transcriptome metabolism such as alternative splicing of nascent messenger RNA

molecules, can be observed allowing for greater depth of understanding of transcriptome

responses. Thus, care must be taken not to assume that even high-density microarrays

can provide a complete picture of the transcriptome.

Table 4.1. Top 20 transcripts induced by TNT exposure

Control TNT TNT/Control SAGE ranka Gene I.D. Annotation

1379 12631 9.16 97 At3g44300 nitrilase 2 572 5212 9.11 NPb At1g54090 hypothetical protein

136572 1105662 8.1 NP At1g17180 glutathione transferase, 133914 894671 6.68 487 At1g02920 glutathione transferase, putative

575 3818 6.63 20 At1g76680 12-oxophytodienoate reductase (OPR1) 60806 389368 6.4 16 At3g21720 putative isocitrate lyase 784 4510 5.75 NP At4g34138 similar to glucosyltransferase -like protein 1135 5744 5.06 NP At1g72900 disease resistance protein (TIR-NBS class),

putative 900 4149 4.61 NP At1g26970 protein kinase, putative

146471 665160 4.54 NP At3g15450 expressed protein 879 3958 4.5 110 At1g02850 glycosyl hydrolase family 1

229008 1004476 4.39 NP At2g29420 glutathione transferase 53984 222163 4.12 NP At1g21000 expressed protein 156249 614698 3.93 6 At5g13750 transporter-like protein 208573 814582 3.91 NP At3g50970 dehydrin Xero2 221705 843792 3.81 247 At4g11600 glutathione peroxidase 323149 1189150 3.68 489 At1g78380 glutathione transferase, putative; 49920 183112 3.67 17 At2g29490 glutathione transferase, 363743 1314001 3.61 1014 At1g02930 glutathione transferase, putative

783 2808 3.58 NP At1g53850 20S proteasome alpha subunit E1 (PAE1) a Ranked according to degree of induction (i.e., descending from most highly induced) bNP=not present in SAGE libraries.

Table 4.2. Top 20 Transcripts Repressed by TNT Exposure

Control TNT Control/TNT SAGE ranka Gene I.D. Annotation

580557 66334 8.75 16 At4g12520 pEARLI 1-like protein 345396 46140 7.49 NPb At5g58850 myb family transcription factor; protein id:

At5g58850.1 23128 3952 5.85 *********** b/t hypothetical protein; protein id: At4g12540.1

and putative cell wall-plasma membrane disconnecting CLCT protein (AIR1A) ; protein id: At4g12550.1

1022035 196517 5.20 NP At2g23830 unknown protein 20234 4309 4.70 NP At3g20950 cytochrome P450, putative 38607 7978 4.84 NP At5g47990 cytochrome p450 family

797955 172184 4.63 7 At2g36830 putative aquaporin (tonoplast intrinsic protein gamma)

4480641 1032280 4.34 NP At5g57660 putative CONSTANS B-box zinc finger protein 3379 804 4.20 NP At3g02620.1 putative stearoyl-acyl carrier protein desaturase 17168 4372 3.93 NP At3g29360 UDP-glucose dehydrogenase, putative

127380 31517 4.04 NP At4g02270 hypothetical protein similar to extensin-like protein

1628890 431025 3.78 NP At5g10430 arabinogalactan-protein (AGP4) 952744 251759 3.78 NP At2g29750 UDP-glycosyltransferase family 5390606 1407372 3.83 6264 At5g44550 putative protein 1220910 314425 3.88 NP At5g21950 similar to putative hydrolase 3137398 805884 3.89 1170 At1g14450 expressed protein 1981036 553714 3.58 1121 At3g26520 gamma tonoplast intrinsic protein 65535 19254 3.40 NP At5g06640 putative protein

973194 280122 3.47 2 At4g12550 putative cell wall-plasma membrane disconnecting CLCT protein (AIR1A)

6476741 1973671 3.28 4025 At1g26250 unknown protein a Ranked according to degree of repression (i.e.,x descending from most highly repressed) bNP=not present in SAGE libraries.

Table 4.3. Comparison of the 20 Most Highly Induced Transcripts as Measured Using SAGE and Microarrays*.

Microarray SAGE

Fold Gene I.D. Annotation Fold Gene I.D. Annotation 9.16 At3g44300 nitrilase 2 27.5 At1g17170 putative glutathione transferase 9.11 At1g54090 hypothetical protein 24.0 At5g11510 MYB like protein

8.10 At1g17180 glutathione transferase*, 23.0 At3g27740

carbamoyl phosphate synthetase small subunit (AGI); pyrimidine nucleotide

biosynthesis--NOT IN 3' MOST EXON

6.68 At1g02920 glutathione transferase, putative 21.0 At5g13750 transporter-like protein

6.63 At1g76680 12-oxophytodienoate reductase (OPR1) 20.0 At1g60940 putative serine/threonine-protein kinase

6.40 At3g21720 putative isocitrate lyase 17.5 At5g03630 monodehydroascorbate reductase (NADH) - like protein

5.75 At4g34138 similar to

glucosyltransferase -like protein

17.0 2 hits

5.06 At1g72900 disease resistance protein

(TIR-NBS class), putative

17.0 At5g63790 putative protein; NAM, no apical meristem, - like protein A.thaliana

4.61 At1g26970 protein kinase, putative 15.0 At1g05050 putative transcription factor

4.54 At3g15450 expressed protein 14.0 At3g28740 cytochrome P450, putative; CYP81D11-A-TYPE

4.50 At1g02850 glycosyl hydrolase family 1 13.0 polyA??

does not appear in the genome (the truncate w/o A's does not match any

putattive genes either) 4.39 At2g29420 glutathione transferase 12.7 At5g45020 putative protein--glutathione transferase 4.12 At1g21000 expressed protein 12.0 one hit on chromosome 3

3.93 At5g13750 transporter-like protein 11.5 At3g21720 putative isocitrate lyase; C-compound and carbohydrate utilization

3.91 At3g50970 dehydrin Xero2 11.0 At2g29490 putative glutathione S-transferase

3.81 At4g11600 glutathione peroxidase 10.0 At1g78080 putative AP2 domain containing protein

RAP2--transcription factor TINY A.thaliana

3.68 At1g78380 glutathione transferase, putative; 10.0 At2g28570 unknown protein; Protein-tyrosine kinase

3.67 At2g29490 glutathione transferase, 10.0 At1g76680 12-oxophytodienoate reductase (OPR1)

3.61 At1g02930 glutathione transferase, putative 10.0 At1g75270 GSH-dependent dehydroascorbate

reductase 1, putative

3.58 At1g53850 20S proteasome alpha subunit E1 (PAE1) 10.0 At1g05680 putative indole-3-acetate beta-

glucosyltransferase

At4g01870; protein of unknown function

*Genes observed in both the microarray analysis and SAGE above are indicated in bold.

Table 4.4. Comparison of the 20 Most Highly Repressed Transcripts as Measured Using SAGE and Microarrays*.

Microarray SAGE

Fold Gene I.D. Annotation Fold Gene I.D. Annotation 8.75 At4g12520 pEARLI 1-like protein 85 At4g12550 putative cell wall-plasma

membrane disconnecting CLCT protein (AIR1A)

7.49 At5g58850 myb family transcription factor

52 At2g01520 major latex protein homolog-like; A.thaliana

5.85 xxxx putative cell wall-plasma membrane disconnecting CLCT protein (AIR1A)

45 At1g70850 major latex protein homolog-like: A. thaliana

5.20 At2g23830 unknown protein 37 AF098631 putative cell wall-plasma membrane disconnecting CLCT

protein (AIR1B) 4.84 At5g47990 cytochrome p450 family 29 At1g09690 putative 60S ribosomal protein L21 4.70 At3g20950 cytochrome P450, putative 29 At2g36830 putative aquaporin (tonoplast

intrinsic protein gamma) 4.63 At2g36830 putative aquaporin

(tonoplast intrinsic protein gamma)

50 At4g22212 Expressed protein

4.34 At5g57660 putative CONSTANS B-box zinc finger protein

24 At5g23710 unknown protein

4.20 At3g02620.1

putative stearoyl-acyl carrier protein desaturase

23 At3g54580 extensin precursor -like protein

4.04 At4g02270 hypothetical protein similar to extensin-like

protein

21 At5g19510 elongation factor 1B alpha-subunit

3.93 At3g29360 UDP-glucose dehydrogenase, putative

20 At5g56540 arabinogalactan-protein AGP14

3.89 At1g14450 expressed protein 20 At1g14960 major latex protein, putative; stress response

3.88 At5g21950 similar to putative hydrolase

20 At4g38740 peptidylprolyl isomerase ROC1; protein folding and stabilization

3.83 At5g44550 putative protein 16 At4g17340 membrane channel like protein; other transport facilitators

3.78 At2g29750 UDP-glycosyltransferase family

15 At4g12520 pEARLI 1-like protein

3.78 At5g10430 arabinogalactan-protein (AGP4)

15 At5g10280 putative transcription factor MYB92

3.58 At3g26520 gamma tonoplast intrinsic protein

56 At5g26260 putative protein;UNCLASSIFIED PROTEINS

3.47 At4g12550 putative cell wall-plasma membrane disconnecting CLCT protein (AIR1A)

42 At5g14200 3-isopropylmalate dehydrogenase; C-compound and carbohydrate

metabolism 3.40 At5g06640 putative protein 14 At4g14960 tubulin alpha-6 chain (TUA6);

cytoskeleton 3.38 At3g02610.

1 putative stearoyl-acyl

carrier protein desaturase 14 At1g04820 tubulin alpha-2/alpha-4 chain;

cytoskeleton *Genes observed in both the microarray analysis and SAGE above are indicated in bold.

REFERENCES

Brown, P.O., and Botstein, D. (1999). Exploring the new world of the genome with

DNA microarrays. Nat. Genet. 21, 33-37.

Deyholos, M.K., and Galbraith, D.W. (2001). High-density DNA microarrays for gene

expression analysis. Cytometry 43, 229-238.

Ishii, M., Hashimoto, S., Tsutsumi, S., Wada, Y., Matsushima, K., Kodama, T., and

Aburatani, H. (2000). Direct comparison of GeneChip and SAGE on the

quantitative accuracy in transcript profiling analysis. Genomics 68, 136-143.

Lashkari, D.A., DeRisi, J.L., McCusker, J.H., Namath, A.F., Gentile, C., Hwang,

S.Y., Brown, P.O., and Davis, R.W. (1997). Yeast microarrays for genome wide

parallel genetic and gene expression analysis. Proc. Natl. Acad. Sci. USA 94,

13057-13062.

Lee, M.L.T., Kuo, F.C., Whitmore, G.A., and Sklar, J. (2000). Importance of

replication in microarray gene expression studies: statistical methods and

evidence form repetitive cDNA hybridizations. Proc. Natl. Acad. Sci. USA 97,

9834-9839.

Lipshutz, R., Fodor, S., Gingeras, T., and Lockhart, D. (1999). High density synthetic

oligonucleotide arrays. Nat. Genet. 21, 20-24.

McGall, G., Labadie, J., Brock, P., Wallraff, G., and Nguyen, T. (1996). Light-

directed synthesis of high-density oligonucleotide arrays using semiconductor

photoresists. Proc. Natl. Acad. Sci. USA 93, 13555-13560.

Nadon, R., and Shoemaker, J. (2002). Statistical issues with microarrays: processing

and analysis. Trends in Genetics 18, 265-271.

Normanly, J., Grisafi, P., Fink, G.R., and Bartel, B. (1997). Arabidopsis mutants

resistant to the auxin effects of indole-3-acetonitrile are defective in the nitrilase

encoded by the NIT1 gene. Plant Cell 9, 1781-1790.

CHAPTER 5

CONCLUSIONS

The viability of phytoremediation as an approach to reducing environmental

contamination is still somewhat in question. While there is a dire need of affordable

methods of remediation--the world market for remediation was estimated at $15-$18

billion and growing in 1998 (Glass, 1999)--several hurdles prevent phytoremediation

from supplanting classical methods of clean up. One major obstacle is the lengthy time

that is often required for complete phytoremediation. Site remediation using plants can

take years, during which time sites must be monitored with respect to contaminant levels

and the trees or plants in use may require care and/or harvesting (Watanabe, 2001).

Furthermore, sites containing high levels of contaminants may inhibit growth and prevent

successful introduction of plants. There are also concerns over the potential for

introduction of contaminants into the food chain. The question of how to dispose of

plants that extract heavy metals or immutable organics is also a consideration.

For these reasons, researchers have endeavored to engineer wild-type plants with

genes that can confer greater tolerance and/or degradation abilities (Rugh et al., 1996;

Rugh et al., 1998; Bizily et al., 1999; French et al., 1999; Hannink et al., 2001).

Although effective under controlled conditions in the laboratory, engineered plants

remain unproven in the field, and some of the species used are unsuitable for field

applications. Furthermore, although less burdened by regulatory requirements than

microbial-based remediation processes, phytoremediation using transgenic or genetically-

modified (GM) plants faces significant public and government opposition. Major

concerns over field release of such plants include increased invasiveness and decreased

genetic variability of native plants due to interbreeding.

Despite these misgivings, researchers continue to pursue the development of

genetically-engineered plants for phytoremediation. Intense interest by federal agencies

and commercial institutions that control numerous sites in need of remediation (there are

about 12,000 contaminated “Superfund” sites listed in the U.S., and 400,000 sites in

Western Europe (Glass, 1999)) is keeping the technology at the forefront, and many

argue that solutions to public concerns exist. For example, the use of sterile clones has

been suggested as a solution to the problem of invasiveness and interbreeding.

As researchers become more adept at introducing genes into a wider variety of

plant species, additional properties desirable for remediation processes such as rapid

growth, deep root structures, and high water uptake, will be available in modified plants

(Dietz and Schnoor, 2001). Yet, there remains a general lack of knowledge with regard to

the metabolic mechanisms used by plants to cope with toxins, which prevents focused

engineering attempts. It has been argued that “increased understanding of the enzymatic

processes involved in plant tolerance and metabolism of xenobiotic chemicals will

provide new potential for engineering plants with increased phytoremediation

capabilities” (Dietz and Schnoor, 2001), but the lack of enzymological information for

many contaminants prevents informed decisions on which genes to engineer. This was

certainly the case for explosives remediation, and it was for this reason that the project

described in this dissertation was undertaken.

Using genomic techniques to measure changes in gene expression in Arabidopsis

thaliana upon exposure to toxic concentrations of TNT and RDX has proven useful for

identifying some of the metabolic mechanisms plants may use for detoxifying explosives.

For example, it appears as though conjugation via glutathione is the major means by

which Arabidopsis deals with the toxic effects of TNT. In addition, the induction of

several cytochrome P450 transcripts reveals the potential for transformation prior to

conjugation. The five-fold increase of a transcript representing an ABC transporter

protein hints at a possible mechanism for conjugate sequestration in keeping with that

observed for herbicide metabolism (Martinoia et al., 1993). Using SAGE, specific gene

family members involved in TNT and RDX metabolism were identified. This

information will be of great utility in the design of transgenic plants for

phytoremediation. While TNT appears to stimulate the function of a multi-phase

mechanism for detoxification in Arabidopsis, RDX appears to rely on a completely

different detoxification mechanism. As noted above, several of the most highly induced

genes in the TNT SAGE library were not even expressed in the presence of RDX.

Moreover, while an aquaporin transcript was highly repressed in the TNT library, another

transcript representing aquaporin was highly induced in the RDX library. This may

indicate different mechanisms for vacuolar metabolism or different strategies for

affecting munitions uptake. These differences will require the engineering of multiple

enhanced pathways for the metabolism of both compounds.

There are also limits to the application of these data. As it is transcriptionally

based, it is also limited in scope. It is well known that transcript levels are not always

related directly to downstream protein levels or enzyme activities. Therefore, measures

of protein expression and activity for genes that appear to be differentially expressed

should be pursued before substantial efforts are expended on creation of transgenic

plants. Knockouts of genes suspected to be involved in the metabolism of TNT and/or

RDX, created using RNA interference (RNAi) or antisense technologies, will aid in

confirming these findings. Studies conducted under field conditions (e.g., growth in

contaminated soil with concentrations of explosives more similar to those at

contaminated sites) are also necessary to better assess the critical physiological changes

induced by exposure to explosives. Finally, transgenic plants with increased or decreased

expression of relevant genes will be necessary to fully test hypotheses arising from the

research presented in this dissertation.

Further studies using transcriptome profiling to fully elucidate plant responses to

explosives will also be necessary. One way to accelerate this process is to combine the

strengths of both techniques. When studying specific processes or responses in which the

genes involved have already been discovered using SAGE or other open-ended

techniques, microarrays provide a powerful method for rapid analysis of cellular

responses. This approach has been used effectively in the study of breast cancer to

elucidate major changes in metabolic pathways during its progression (Nacht et al.,

1999). Using SAGE, researchers identified genes that were differentially expressed and

then used these candidates to fabricate custom microarrays for screening clinical breast

tumor samples to find genes that are consistently expressed at different levels in diseased

and normal tissues.

Custom microarrays have also been used for rapid assessment of drug or pollutant

mechanisms of toxicity (Nuwaysir et al., 1999). This approach can be an extremely

efficient means of screening drug candidates for toxicity prior to conducting more costly

and time-consuming studies. In environmental research, identification of genes

responding to toxicant exposure can be useful in risk assessment. If a compound whose

mechanism of action is unknown induces expression of genes previously shown to be

induced by exposure to a class of toxicant whose mechanism of action has already been

elucidated (e.g., endocrine disruptors), the unknown compound’s toxicity may be rapidly

classified and its threat to the environment more accurately assessed (Nuwaysir, 1999)

(Figure 5.1). While this method has been employed in human toxicity studies (e.g.,

ToxChip v1.0 developed by Nuwaysir and colleagues), it has yet to be employed in the

study of plant responses (Nuwaysir et al., 1999). It will be useful for future toxicological

studies to develop a similar microarray or “ToxChip” for Arabidopsis.

While microbial bioremediation has already achieved moderate success,

phytoremediation is still mostly in the proving phase. Although it has shown great

promise in the laboratory, there are relatively few examples of its successful transition

into the field. However, as this new technology develops, the limitations responsible for

the delay in its successful application will be overcome. The ecological benefits afforded

by phytoremediation provide the impetus for pursuing its widespread implementation.

The use of “natural” solutions, such as phytoremediation, encourages the reestablishment

of biological systems on contaminated sites, while current methods

(excavation/incineration) do just the opposite. Thus, public acceptance of this technology

as an alternative to more destructive processes will help to bring it to the forefront of

remediation strategies.

Figure 5.1. Schematic representation of a microarray-based method for

identifying a toxicant mechanism of action. In this method, gene-expression data derived

from exposure of a model system to known toxicants is compared to changes induced by

novel compounds (termed the toxicant signature) (Nuwaysir, 1999).

REFERENCES

Bizily, S., Rugh, C.L., Summer, A.O., and Meagher, R.B. (1999). Phytoremediation of

methylmercury pollution: merB expression in Arabidopsis thaliana confers

resistance to organomercurials. Proc. Natl. Acad. Sci. USA 96, 6808-6813.

Dietz, A.C., and Schnoor, J.L. (2001). Advances in phytoremediation. Environ. Health

Persp. 109, 163-168.

French, C.E., Rosser, S.J., Davies, G.J., Nicklin, S., and Bruce, N.C. (1999).

Biodegradation of explosives by transgenic plants expression pentaerythritol

tetranitrate reductase. Nat. Biotechnol. 17, 491-494.

Glass, D. (1999). U.S. and international markets for phytoremediation. (D. Glass

Associates, Inc.; 1999).

http://www.channel1.com/users/dglass/INFO/phy99exc.htm.

Hannink, N., Rosser, S.J., French, C.E., Basran, A., Murray, J.A.H., Nicklin, S., and

Bruce, N.C. (2001). Phytodetoxification of TNT by transgenic plants expressing

a bacterial nitroreductase. Nat. Biotechnol. 19, 1168-1172.

Martinoia, E., Grill, E., Tommasini, R., Kreuz, K., and Amrhein, M. (1993). An

ATP-dependent glutathione-S-conjugate 'export' pump in the vacuolar membrane

of plants. Nature 364, 247-249.

Nacht, M., Ferguson, A.T., Zhang, W., Petroziello, J.M., Cook, B.P., Gao, Y.H.,

Maguire, S., Riley, D., Coppola, G., Landes, G.M., Madden, S.L., and

Sukumar, S. (1999). Combining serial analysis of gene expression and array

technologies to identify genes differentially expressed in breast cancer. Cancer

Res. 59, 5464-5470.

Nuwaysir, E.F., Bittner, M., Trent, J., Barrett, J.C., and Afshari, C.A. (1999).

Microarrays and toxicology: the advent of toxicogenomics. Mol. Carcin. 24, 153-

Rugh, C.L., Senecoff, J.F., Meagher, R.B., and Merkle, S.A. (1998). Development of

transgenic yellow poplars for mercury phytoremediation. Nat. Biotechnol. 16,

925-928.

Rugh, C.L., Wilde, D., Stack, N.M., Thompson, D.M., Summer, A.O., and Meagher,

R.B. (1996). Mercuric ion reduction and resistance in transgenic Arabidopsis

thaliana plants expressing a modified bacterial merA gene. Proc. Natl. Acad. Sci.

USA 93, 3182-3187.

Watanabe, M.E. (2001). Can bioremediation bounce back? Nat. Biotechnol. 19, 1111-

transcriptional profiling of arabidopsis thaliana

Documents

a transcriptional dynamic network during arabidopsis...

research article transcriptional consequence and...

whole genome transcriptome variation in arabidopsis thaliana...

characterizing auxin biosynthetic mutants in arabidopsis...

arabidopsis thaliana inspired genetic restoration strategies

arabidopsis thaliana - hindawi

arabidopsis thaliana genome project

restriction digestion of arabidopsis thaliana genomic dna

the multinational coordinated arabidopsis thaliana ... ·...

the multinational coordinated arabidopsis thaliana

zinc uptake and storage in arabidopsis thaliana

whole genome sequencing of arabidopsis thaliana

stable transgenic arabidopsis thaliana

telomere-associated proteins in arabidopsis thaliana

arabidopsis thaliana transformation by agrobacterium ... ·...

the dna dioxygenase alkbh2 protects arabidopsis thaliana

v6 circadian clocks in arabidopsis thaliana

gwas in a model organism: arabidopsis thaliana

restriction digestion of arabidopsis thaliana genomic dna

protein expression in arabidopsis thaliana › archive ›...